content
large_stringlengths
3
20.5k
url
large_stringlengths
53
192
branch
large_stringclasses
4 values
source
large_stringclasses
51 values
embeddings
listlengths
384
384
score
float64
-0.21
0.65
`: Gateway token (if required) - `--password `: Gateway password (if required) - `--session `: Session key (default: `main`, or `global` when scope is global) - `--deliver`: Deliver assistant replies to the provider (default off) - `--thinking `: Override thinking level for sends - `--message `: Send an initial message...
https://github.com/openclaw/openclaw/blob/main/docs/web/tui.md
main
opebclaw
[ -0.06394697725772858, -0.030474577099084854, -0.06768234074115753, 0.03475505858659744, -0.04561981186270714, -0.03368134796619415, -0.01911093480885029, -0.00414508581161499, -0.030681638047099113, 0.06338652968406677, 0.003839587327092886, -0.023186152800917625, 0.0024304327089339495, 0....
0.083842
The Gateway dashboard is the browser Control UI served at `/` by default (override with `gateway.controlUi.basePath`). Quick open (local Gateway): - [http://127.0.0.1:18789/](http://127.0.0.1:18789/) (or [http://localhost:18789/](http://localhost:18789/)) - With `gateway.tls.enabled: true`, use `https://127.0.0.1:18789...
https://github.com/openclaw/openclaw/blob/main/docs/web/dashboard.md
main
opebclaw
[ -0.06607641279697418, 0.006379157304763794, -0.0524396188557148, 0.006893039681017399, -0.023452404886484146, -0.060865070670843124, -0.00016341175069101155, 0.049520041793584824, 0.05921055004000664, -0.034720733761787415, 0.01773574948310852, 0.009907812811434269, 0.012204793281853199, 0...
0.035163
the dashboard's requested scopes; re-pair or approve the requested scope contract instead of rotating the shared gateway token. - Outside that retry path, connect auth precedence is explicit shared token/password first, then explicit `deviceToken`, then stored device token, then bootstrap token. - On the async Tailscal...
https://github.com/openclaw/openclaw/blob/main/docs/web/dashboard.md
main
opebclaw
[ -0.06580278277397156, -0.003967730328440666, 0.023941628634929657, 0.07042834907770157, 0.02737371250987053, -0.0635460913181305, -0.021581383422017097, 0.010426318272948265, 0.02650376595556736, -0.007953980937600136, 0.03534403443336487, -0.01625831425189972, 0.061422623693943024, -0.004...
0.005063
Status: the macOS/iOS SwiftUI chat UI talks directly to the Gateway WebSocket. ## What it is - A native chat UI for the gateway (no embedded browser and no local static server). - Uses the same sessions and routing rules as other channels. - Deterministic routing: replies always go back to WebChat. ## Quick start 1. St...
https://github.com/openclaw/openclaw/blob/main/docs/web/webchat.md
main
opebclaw
[ -0.035296421498060226, -0.07536200433969498, 0.04515348747372627, 0.028691688552498817, -0.045582011342048645, -0.08786185830831528, 0.034807272255420685, -0.03353559970855713, 0.028019405901432037, -0.018627846613526344, 0.015464531257748604, 0.019167203456163406, -0.07067982107400894, 0....
0.088636
WebChat does not write arbitrary delivery, status, or helper text into that transcript. - Gateway `ReplyPayload` events are the live delivery projection. They can be normalized for WebChat/channel display, block streaming, directive tags, media embedding, TTS/audio flags, and UI fallback behavior. They are not themselv...
https://github.com/openclaw/openclaw/blob/main/docs/web/webchat.md
main
opebclaw
[ -0.024113787338137627, -0.0725097730755806, 0.029915479943156242, 0.024396387860178947, 0.036099329590797424, -0.0796804204583168, 0.009302452206611633, -0.023234648630023003, 0.0390055887401104, 0.0035965675488114357, 0.03989725559949875, -0.007668933365494013, -0.05646200105547905, 0.040...
0.061831
The Gateway serves a small \*\*browser Control UI\*\* (Vite + Lit) from the same port as the Gateway WebSocket: - default: `http://:18789/` - with `gateway.tls.enabled: true`: `https://:18789/` - optional prefix: set `gateway.controlUi.basePath` (e.g. `/openclaw`) Capabilities live in [Control UI](/web/control-ui). The...
https://github.com/openclaw/openclaw/blob/main/docs/web/index.md
main
opebclaw
[ -0.08101027458906174, 0.011616000905632973, -0.12350340932607651, 0.04997538775205612, -0.01580236479640007, -0.0507592111825943, 0.00004941569932270795, 0.030545057728886604, 0.02952101267874241, -0.03498132526874542, 0.03685978427529335, 0.021652821451425552, -0.02949695661664009, -0.019...
0.138955
The Control UI is a small \*\*Vite + Lit\*\* single-page app served by the Gateway: - default: `http://:18789/` - optional prefix: set `gateway.controlUi.basePath` (e.g. `/openclaw`) It speaks \*\*directly to the Gateway WebSocket\*\* on the same port. ## Quick open (local) If the Gateway is running on the same compute...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.07640278339385986, -0.004379235208034515, -0.055139582604169846, 0.052618566900491714, -0.023820795118808746, -0.05384188890457153, -0.06071503832936287, 0.02736702747642994, -0.009031642228364944, -0.05355573445558548, 0.047520849853754044, -0.010871143080294132, -0.01347245741635561, ...
0.033409
directly (such as scripted gateways or custom dashboards). ## Runtime config endpoint The Control UI fetches its runtime settings from `/\_\_openclaw/control-ui-config.json`. That endpoint is gated by the same gateway auth as the rest of the HTTP surface: unauthenticated browsers cannot fetch it, and a successful fetch...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.03557772561907768, -0.02097022533416748, -0.07686597853899002, 0.035245463252067566, -0.04950613155961037, -0.04294033721089363, -0.02642776444554329, 0.019412336871027946, 0.015314217656850815, -0.007271763868629932, 0.01158051285892725, -0.05082671344280243, -0.015389853157103062, 0.0...
0.045024
while slow provider checks finish, and partial snapshots are labeled when a probe or audit exceeds its UI budget. - Instances: presence list + refresh (`system-presence`). - Sessions: list configured-agent sessions by default, fall back from stale unconfigured agent session keys, and apply per-session model/thinking/fa...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ 0.00006725954881403595, 0.018745213747024536, -0.029710570350289345, 0.06076005473732948, 0.07404925674200058, -0.04891221225261688, 0.018669219687581062, -0.02997644804418087, -0.02657417394220829, 0.018758097663521767, 0.03276901692152023, -0.004519506357610226, -0.025123635306954384, -0...
0.111595
by itself; use it to inspect and edit saved config, then use `openclaw mcp doctor --probe` when you need live server proof. Typical workflow: 1. Open \*\*MCP\*\* from the sidebar. 2. Check the summary cards for total, enabled, OAuth, and filtered server counts. 3. Review each server row for transport, enablement, auth,...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.026861462742090225, 0.046053607016801834, -0.0220043808221817, 0.01893145591020584, 0.03752380982041359, -0.06445811688899994, -0.062450654804706573, -0.009757384657859802, -0.04921813681721687, -0.00890953280031681, -0.004168835934251547, 0.014826986007392406, 0.027864456176757812, -0....
0.012258
heartbeat acknowledgement token `HEARTBEAT\_OK`. - During an active send and the final history refresh, the chat view keeps local optimistic user/assistant messages visible if `chat.history` briefly returns an older snapshot; the canonical transcript replaces those local messages once the Gateway history catches up. - ...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.07336867600679398, -0.00858487468212843, 0.044887952506542206, 0.003910106141120195, 0.0348975770175457, -0.04203367605805397, 0.04202154278755188, -0.08017437905073166, 0.08043093979358673, -0.01389184407889843, 0.010383570566773415, 0.010078972205519676, -0.07712425291538239, -0.00535...
0.133974
stay server-side while browser audio moves through authenticated Gateway RPCs. The Realtime session prompt is assembled by the Gateway; `talk.client.create` does not accept caller-provided instruction overrides. The Chat composer includes a Talk options button next to the Talk start/stop button. The options apply to th...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.01588387042284012, -0.0665833055973053, -0.05531429126858711, 0.04551476985216141, 0.000006011140612827148, -0.008128858171403408, 0.0014658744912594557, -0.08320624381303787, 0.07762619853019714, -0.023685690015554428, -0.001202341984026134, 0.06726973503828049, -0.09206706285476685, -...
0.039928
an `endpoint` plus `keys.p256dh`/`keys.auth`. - `push.web.unsubscribe` — removes a registered endpoint. - `push.web.test` — sends a test notification to the caller's subscription. Web Push is independent of the iOS APNS relay path (see [Configuration](/gateway/configuration) for relay-backed push) and the existing `pus...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.053567562252283096, -0.03241478279232979, 0.04568580165505409, 0.0016200491227209568, 0.03842894732952118, -0.013088775798678398, -0.0569155216217041, -0.05079163238406181, 0.025316830724477768, -0.010093479417264462, 0.04624372720718384, 0.06744647771120071, -0.007590354420244694, 0.03...
0.08602
"tailnet", auth: { mode: "token", token: "replace-me" }, }, } ``` `allowInsecureAuth` is a local compatibility toggle only: - It allows localhost Control UI sessions to proceed without device identity in non-secure HTTP contexts. - It does not bypass pairing checks. - It does not relax remote (non-localhost) device ide...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.05823911726474762, 0.04943384975194931, 0.03079414740204811, -0.007676669396460056, 0.04302519932389259, -0.03958169370889664, 0.024022212252020836, -0.02025436796247959, -0.033235419541597366, 0.021888552233576775, 0.019374096766114235, 0.02409176714718342, 0.017520129680633545, 0.0605...
0.032797
server): ```bash pnpm ui:dev ``` Then point the UI at your Gateway WS URL (e.g. `ws://127.0.0.1:18789`). ## Blank Control UI page If the browser loads a blank dashboard and DevTools shows no useful error, an extension or early content script may have prevented the JavaScript module app from evaluating. The static page ...
https://github.com/openclaw/openclaw/blob/main/docs/web/control-ui.md
main
opebclaw
[ -0.006911449134349823, -0.047993507236242294, -0.006989450193941593, -0.0183890238404274, 0.06001449376344681, -0.04181188717484474, -0.07230491936206818, 0.043634504079818726, 0.006485081277787685, 0.03506571426987648, 0.05833042412996292, 0.025458335876464844, -0.0703539103269577, 0.0070...
-0.022246
[DeepSeek](https://www.deepseek.com) provides powerful AI models with an OpenAI-compatible API. | Property | Value | | -------- | -------------------------- | | Provider | `deepseek` | | Auth | `DEEPSEEK\_API\_KEY` | | API | OpenAI-compatible | | Base URL | `https://api.deepseek.com` | ## Getting started Create an API ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/deepseek.md
main
opebclaw
[ -0.0385231077671051, -0.08050815761089325, -0.07189907878637314, 0.050576575100421906, 0.026855818927288055, -0.08295948803424835, -0.11315656453371048, 0.019718382507562637, -0.05267955735325813, -0.04584062844514847, 0.016940006986260414, -0.06118442863225937, -0.006015310995280743, 0.00...
0.113709
[Hugging Face Inference Providers](https://huggingface.co/docs/inference-providers) offer OpenAI-compatible chat completions through a single router API. You get access to many models (DeepSeek, Llama, and more) with one token. OpenClaw uses the \*\*OpenAI-compatible endpoint\*\* (chat completions only); for text-to-im...
https://github.com/openclaw/openclaw/blob/main/docs/providers/huggingface.md
main
opebclaw
[ -0.11233842372894287, -0.004822596907615662, 0.003212075214833021, 0.04049446061253548, 0.08513009548187256, -0.07859625667333603, -0.039573706686496735, -0.007760514970868826, 0.03282904252409935, -0.021468443796038628, 0.015619573183357716, -0.09862494468688965, -0.010518087074160576, 0....
0.125367
can set a custom label per model in config so it appears the way you want in the CLI and UI: ```json5 { agents: { defaults: { models: { "huggingface/deepseek-ai/DeepSeek-R1": { alias: "DeepSeek R1 (fast)" }, "huggingface/deepseek-ai/DeepSeek-R1:cheapest": { alias: "DeepSeek R1 (cheap)" }, }, }, }, } ``` - \*\*Policy su...
https://github.com/openclaw/openclaw/blob/main/docs/providers/huggingface.md
main
opebclaw
[ -0.08159829676151276, 0.02029428817331791, -0.004612659104168415, 0.03555809333920479, -0.01924016699194908, -0.005148217082023621, -0.04956468939781189, 0.02682182937860489, -0.0027272857259958982, -0.06446826457977295, 0.03188315033912659, -0.09381598234176636, 0.013370941393077374, 0.04...
0.062171
OpenClaw now treats Qwen as a first-class bundled provider with canonical id `qwen`. The bundled provider targets the Qwen Cloud / Alibaba DashScope and Coding Plan endpoints, keeps legacy `modelstudio` ids working as a compatibility alias, and also exposes the Qwen Portal token flow as provider `qwen-oauth`. - Provide...
https://github.com/openclaw/openclaw/blob/main/docs/providers/qwen.md
main
opebclaw
[ -0.09488627314567566, -0.01685193181037903, -0.01997290924191475, -0.011600200086832047, -0.005817376542836428, -0.018418246880173683, -0.030361073091626167, -0.009758648462593555, 0.04430034011602402, 0.017001118510961533, 0.05103670060634613, -0.03514682129025459, 0.054338712245225906, 0...
-0.030249
| Model ref | Input | Context | Notes | | --------------------------- | ----------- | --------- | -------------------------------------------------- | | `qwen/qwen3.5-plus` | text, image | 1,000,000 | Default model | | `qwen/qwen3.6-plus` | text, image | 1,000,000 | Prefer Standard endpoints when you need this model | ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/qwen.md
main
opebclaw
[ -0.06908728182315826, -0.002920308383181691, -0.04668916389346123, -0.022166496142745018, -0.035577353090047836, 0.05047554150223732, -0.006432391703128815, 0.07002271711826324, 0.04259521886706352, 0.009357331320643425, 0.07586953788995743, -0.09570685774087906, 0.052394166588783264, -0.0...
-0.008974
output video per request - Up to \*\*1\*\* input image - Up to \*\*4\*\* input videos - Up to \*\*10 seconds\*\* duration - Supports `size`, `aspectRatio`, `resolution`, `audio`, and `watermark` - Reference image/video mode currently requires \*\*remote http(s) URLs\*\*. Local file paths are rejected up front because t...
https://github.com/openclaw/openclaw/blob/main/docs/providers/qwen.md
main
opebclaw
[ 0.012883797287940979, -0.06841414421796799, -0.024186300113797188, -0.021020784974098206, 0.054233916103839874, -0.01882491260766983, -0.050571709871292114, -0.004519619047641754, 0.034431032836437225, 0.015551655553281307, 0.004663258790969849, -0.04448099806904793, -0.04596397280693054, ...
0.032208
OpenClaw ships a bundled `comfy` plugin for workflow-driven ComfyUI runs. The plugin is entirely workflow-driven, so OpenClaw does not try to map generic `size`, `aspectRatio`, `resolution`, `durationSeconds`, or TTS-style controls onto your graph. | Property | Detail | | --------------- | -----------------------------...
https://github.com/openclaw/openclaw/blob/main/docs/providers/comfy.md
main
opebclaw
[ -0.018627332523465157, -0.016115963459014893, -0.01615099608898163, 0.05246353894472122, 0.04972397908568382, -0.05049519240856171, -0.03304721787571907, -0.009991929866373539, 0.0016527881380170584, -0.03717716410756111, 0.022849205881357193, -0.0894462913274765, 0.00746544823050499, 0.00...
0.054707
| Optional inline key, alternative to `COMFY\_API\_KEY` / `COMFY\_CLOUD\_API\_KEY` env vars. | | `allowPrivateNetwork` | boolean | Allow a private/LAN `baseUrl` in cloud mode. | ### Per-capability keys These keys apply inside the `image`, `video`, or `music` sections: | Key | Required | Default | Description | | ------...
https://github.com/openclaw/openclaw/blob/main/docs/providers/comfy.md
main
opebclaw
[ 0.0021228957921266556, -0.006377062294632196, -0.01797201670706272, -0.018972033634781837, 0.02695939503610134, 0.01039578951895237, -0.02664179354906082, -0.05013398453593254, 0.025709984824061394, -0.017223946750164032, 0.06360293179750443, -0.020788026973605156, 0.03607036918401718, 0.0...
0.016443
[Fireworks](https://fireworks.ai) exposes open-weight and routed models through an OpenAI-compatible API. OpenClaw includes a bundled Fireworks provider plugin that ships with two pre-cataloged Kimi models and accepts any Fireworks model or router id at runtime. | Property | Value | | --------------- | ----------------...
https://github.com/openclaw/openclaw/blob/main/docs/providers/fireworks.md
main
opebclaw
[ -0.014907550066709518, -0.01173793338239193, -0.09013432264328003, 0.08109674602746964, 0.07252595573663712, -0.040786877274513245, -0.010593882761895657, -0.011866888031363487, -0.025912988930940628, -0.019035328179597855, -0.004890709184110165, -0.04414447396993637, 0.03037326969206333, ...
0.138788
to that process — not just to your interactive shell. A key exported only in an interactive shell will not help a launchd or systemd daemon unless that environment is imported there too. Set the key in `~/.openclaw/.env` or via `env.shellEnv` to make it readable from the gateway process. On macOS, `openclaw gateway ins...
https://github.com/openclaw/openclaw/blob/main/docs/providers/fireworks.md
main
opebclaw
[ 0.024939535185694695, -0.05728490278124809, -0.008035175502300262, -0.016038300469517708, 0.012048388831317425, -0.06044844910502434, -0.057498104870319366, 0.016934063285589218, -0.05282355099916458, 0.014163563959300518, 0.015113873407244682, 0.005332117434591055, 0.0640537366271019, -0....
-0.005064
OpenClaw ships a bundled `xai` provider plugin for Grok models. For most users, the recommended path is Grok OAuth with an eligible SuperGrok or X Premium subscription. OpenClaw stays local-first: the Gateway, config, routing, and tools run on your machine, while Grok model requests authenticate through xAI and are sen...
https://github.com/openclaw/openclaw/blob/main/docs/providers/xai.md
main
opebclaw
[ -0.03923744708299637, 0.013941168785095215, -0.11447994410991669, -0.053811993449926376, 0.038652803748846054, -0.10559212416410446, -0.034284383058547974, 0.03241247311234474, 0.05597170814871788, -0.05019016191363335, 0.03146011754870415, -0.06730815768241882, 0.054212603718042374, -0.03...
0.01522
for the completed token exchange. ## Built-in catalog OpenClaw includes the current xAI chat models out of the box, ordered newest first in model pickers: | Family | Model ids | | -------------- | ------------------------------------------------------------------------ | | Grok Build 0.1 | `grok-build-0.1` | | Grok 4.3...
https://github.com/openclaw/openclaw/blob/main/docs/providers/xai.md
main
opebclaw
[ -0.08327104896306992, -0.08439937978982925, 0.008694574236869812, -0.002614746568724513, 0.052965741604566574, -0.09775486588478088, -0.0041807726956903934, -0.01009040791541338, 0.02829955704510212, 0.024308349937200546, 0.0749681293964386, -0.058421287685632706, -0.049841783940792084, 0....
0.149096
- Duration: 1-15 seconds for generation/image-to-video, 1-10 seconds when using `reference\_image` roles, 2-10 seconds for extension - Reference-image generation: set `imageRoles` to `reference\_image` for every supplied image; xAI accepts up to 7 such images - Default operation timeout: 600 seconds unless `video\_gene...
https://github.com/openclaw/openclaw/blob/main/docs/providers/xai.md
main
opebclaw
[ 0.01266003679484129, -0.03048843704164028, -0.021058762446045876, 0.00008456569776171818, 0.08061050623655319, -0.06619979441165924, -0.09203337132930756, -0.012457607313990593, 0.07179782539606094, -0.07684150338172913, 0.047817498445510864, -0.05478508397936821, -0.012877194210886955, 0....
0.003599
xAI provider can forward those frames directly without transcoding: ```json5 { plugins: { entries: { "voice-call": { config: { streaming: { enabled: true, provider: "xai", providers: { xai: { apiKey: "${XAI\_API\_KEY}", endpointingMs: 800, language: "en", }, }, }, }, }, }, }, } ``` Provider-owned config lives under `pl...
https://github.com/openclaw/openclaw/blob/main/docs/providers/xai.md
main
opebclaw
[ -0.0615459606051445, -0.1015365868806839, -0.07463591545820236, -0.055316731333732605, 0.0192794818431139, 0.002634976524859667, 0.03693103417754173, -0.03837224468588829, 0.029462292790412903, -0.0598057359457016, 0.06077880412340164, -0.054925449192523956, -0.012636582367122173, 0.054464...
0.092176
tool-schema flags and reasoning payload keys before sending native xAI requests. - `web\_search`, `x\_search`, and `code\_execution` are exposed as OpenClaw tools. OpenClaw enables the specific xAI built-in it needs inside each tool request instead of attaching all native tools to every chat turn. - Grok `web\_search` ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/xai.md
main
opebclaw
[ -0.042541805654764175, -0.008854160085320473, -0.07795530557632446, -0.030200043693184853, 0.05022267997264862, -0.12486313283443451, -0.0010244360892102122, 0.006978181190788746, 0.014785264618694782, -0.036627281457185745, 0.06630277633666992, -0.11278922110795975, 0.03000224195420742, 0...
0.12091
Tencent Cloud ships as a bundled provider plugin in OpenClaw. It gives access to Tencent Hy3 preview through the TokenHub endpoint (`tencent-tokenhub`) using an OpenAI-compatible API. | Property | Value | | ---------------- | ----------------------------------------------------- | | Provider id | `tencent-tokenhub` | |...
https://github.com/openclaw/openclaw/blob/main/docs/providers/tencent.md
main
opebclaw
[ -0.061752915382385254, -0.041886456310749054, -0.07448621839284897, 0.006308132316917181, 0.0735887885093689, 0.01586822047829628, -0.06590615957975388, -0.013761334121227264, -0.02734140306711197, 0.015791233628988266, 0.08196016401052475, -0.070308618247509, 0.025595638900995255, 0.02678...
0.048694
DeepInfra provides a \*\*unified API\*\* that routes requests to the most popular open source and frontier models behind a single endpoint and API key. It is OpenAI-compatible, so most OpenAI SDKs work by switching the base URL. ## Getting an API key 1. Go to [https://deepinfra.com/](https://deepinfra.com/) 2. Sign in ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/deepinfra.md
main
opebclaw
[ -0.06295114010572433, -0.08509854227304459, -0.06267344951629639, -0.0052352845668792725, 0.024183198809623718, -0.09892230480909348, -0.11887039989233017, 0.04956088587641716, -0.013904904946684837, -0.024286959320306778, -0.03938908874988556, -0.05540478974580765, -0.0363147146999836, 0....
0.080309
Anthropic builds the \*\*Claude\*\* model family. OpenClaw supports two auth routes: - \*\*API key\*\* — direct Anthropic API access with usage-based billing (`anthropic/\*` models) - \*\*Claude CLI\*\* — reuse an existing Claude Code login on the same host OpenClaw's Claude CLI backend runs the installed Claude Code C...
https://github.com/openclaw/openclaw/blob/main/docs/providers/anthropic.md
main
opebclaw
[ -0.06889981776475906, 0.027312980964779854, -0.1267162710428238, -0.02348903939127922, -0.023384464904665947, -0.03757119178771973, -0.007707524579018354, 0.005322277545928955, 0.013567352667450905, 0.0327107198536396, 0.003910179249942303, -0.0052703628316521645, 0.012975595891475677, -0....
0.076749
Claude CLI. OpenClaw also supports subscription-style options from [OpenAI Codex](/providers/openai), [Qwen Cloud](/providers/qwen), [MiniMax](/providers/minimax), and [Z.AI / GLM](/providers/zai). ## Thinking defaults (Claude 4.8 and 4.6) Claude Opus 4.8 keeps thinking off by default in OpenClaw. When you explicitly e...
https://github.com/openclaw/openclaw/blob/main/docs/providers/anthropic.md
main
opebclaw
[ -0.04753262922167778, -0.017796697095036507, -0.0644235610961914, 0.03163718804717064, -0.009599928744137287, -0.016590580344200134, -0.06501110643148422, 0.0032789602410048246, 0.04774361476302147, -0.0014634976396337152, -0.005981580354273319, 0.000525331066455692, -0.01025594025850296, ...
0.040253
true` also applies to the Claude CLI backend (`claude-cli/\*`) for eligible GA-capable Opus and Sonnet models, preserving the runtime context window for those CLI sessions to match the direct-API behavior. Requires long-context access on your Anthropic credential. OAuth/subscription token auth keeps its required Anthro...
https://github.com/openclaw/openclaw/blob/main/docs/providers/anthropic.md
main
opebclaw
[ -0.053507283329963684, -0.036966171115636826, -0.11689198762178421, -0.016454817727208138, 0.016212917864322662, -0.09089236706495285, -0.0010168657172471285, -0.034222312271595, 0.022194430232048035, -0.011174241080880165, -0.016953621059656143, -0.03379068523645401, 0.01929572783410549, ...
0.050458
[Groq](https://groq.com) provides ultra-fast inference on open-weight models (Llama, Gemma, Kimi, Qwen, GPT OSS, and more) using custom LPU hardware. OpenClaw includes a bundled Groq plugin that registers both an OpenAI-compatible chat provider and an audio media-understanding provider. | Property | Value | | ---------...
https://github.com/openclaw/openclaw/blob/main/docs/providers/groq.md
main
opebclaw
[ -0.061715107411146164, -0.08328569680452347, -0.06783444434404373, 0.04025213420391083, 0.018467571586370468, -0.03746246173977852, 0.05001106113195419, -0.03823722526431084, -0.016256842762231827, -0.03529135510325432, 0.02408629283308983, -0.07024282217025757, -0.05072614923119545, 0.037...
0.148501
[{ provider: "groq" }], }, }, }, } ``` If the Gateway runs as a managed service (launchd, systemd, Docker), `GROQ\_API\_KEY` must be visible to that process — not just to your interactive shell. A key exported only in an interactive shell will not help a launchd or systemd daemon unless that environment is imported the...
https://github.com/openclaw/openclaw/blob/main/docs/providers/groq.md
main
opebclaw
[ -0.005098119843751192, 0.0101673798635602, -0.08875211328268051, 0.01925739273428917, 0.0177936814725399, -0.032104216516017914, -0.010227547027170658, -0.019140873104333878, 0.02345999702811241, 0.00776113523170352, -0.028065605089068413, -0.006282202433794737, 0.0030301380902528763, 0.02...
0.041542
[LiteLLM](https://litellm.ai) is an open-source LLM gateway that provides a unified API to 100+ model providers. Route OpenClaw through LiteLLM to get centralized cost tracking, logging, and the flexibility to switch backends without changing your OpenClaw config. \*\*Why use LiteLLM with OpenClaw?\*\* - \*\*Cost track...
https://github.com/openclaw/openclaw/blob/main/docs/providers/litellm.md
main
opebclaw
[ -0.0033757230266928673, -0.07554163783788681, -0.0940980613231659, 0.10653901845216751, 0.014841657131910324, -0.10133206099271774, -0.04885692149400711, 0.025622965767979622, -0.05616220086812973, -0.0019081530626863241, -0.003854357637465, -0.006101475562900305, -0.009015475399792194, -0...
0.080219
\*\*claude-max-api-proxy\*\* is a community tool that exposes your Claude Max/Pro subscription as an OpenAI-compatible API endpoint. This allows you to use your subscription with any tool that supports the OpenAI API format. This path is technical compatibility only. Anthropic has blocked some subscription usage outsid...
https://github.com/openclaw/openclaw/blob/main/docs/providers/claude-max-api-proxy.md
main
opebclaw
[ -0.04194864258170128, 0.000998245319351554, -0.11954697966575623, -0.045887574553489685, 0.021666958928108215, 0.009629510343074799, -0.0015087614301592112, 0.053412336856126785, -0.0122007941827178, 0.015872007235884666, -0.02305719442665577, 0.0047343578189611435, -0.0016096264589577913, ...
0.063489
keys, see [Anthropic provider](/providers/anthropic). For OpenAI/Codex subscriptions, see [OpenAI provider](/providers/openai). ## Related Native OpenClaw integration with Claude CLI or API keys. For OpenAI/Codex subscriptions. Overview of all providers, model refs, and failover behavior. Full config reference.
https://github.com/openclaw/openclaw/blob/main/docs/providers/claude-max-api-proxy.md
main
opebclaw
[ -0.06654524803161621, -0.03185928985476494, -0.09960796684026718, 0.018946362659335136, 0.04615345224738121, -0.05934237688779831, 0.005379772279411554, -0.016174547374248505, -0.014983568340539932, -0.01937316171824932, 0.04733291268348694, -0.003294798079878092, 0.013578028418123722, -0....
0.169916
[Gradium](https://gradium.ai) is a bundled text-to-speech provider for OpenClaw. The plugin can render normal audio replies (WAV), voice-note-compatible Opus output, and 8 kHz u-law audio for telephony surfaces. | Property | Value | | ------------- | ------------------------------------ | | Provider id | `gradium` | | ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/gradium.md
main
opebclaw
[ -0.057780034840106964, -0.054204970598220825, -0.15235121548175812, -0.08797517418861389, -0.042900409549474716, -0.023675348609685898, -0.03479619324207306, -0.002834593877196312, 0.020356005057692528, -0.05895098298788071, 0.035714879631996155, -0.05127936601638794, -0.011502685956656933, ...
0.082792
OpenAI provides developer APIs for GPT models, and Codex is also available as a ChatGPT-plan coding agent through OpenAI's Codex clients. OpenClaw uses one provider id, `openai`, for both auth shapes. OpenClaw uses `openai/\*` as the canonical OpenAI model route. Embedded agent turns on OpenAI models run through the na...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.06745753437280655, -0.05859775096178055, -0.10085085779428482, 0.018729494884610176, 0.05175875127315521, -0.07303618639707565, -0.006389758549630642, 0.04360878840088844, 0.05020346865057945, -0.03248068690299988, -0.01942068710923195, 0.0019192746840417385, 0.01963036321103573, 0.0248...
0.127011
agent turns. Use OpenAI API-key profiles only when you want direct API-key auth for an OpenAI agent model. OpenAI agent model turns require the bundled Codex app-server plugin. Explicit OpenClaw runtime config remains available as an opt-in compatibility route. When OpenClaw is explicitly selected with an `openai` OAut...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.021499110385775566, -0.01935388334095478, -0.09443005174398422, 0.03674199432134628, 0.025212425738573074, -0.08951366692781448, -0.0730966180562973, 0.014109472744166851, -0.018905961886048317, -0.03990872576832771, 0.03730315342545509, -0.02517169527709484, 0.02152540162205696, 0.0252...
0.029672
```bash openclaw onboard --openai-api-key "$OPENAI\_API\_KEY" ``` ```bash openclaw models list --provider openai ``` ### Route summary | Model ref | Runtime config | Route | Auth | | ---------------------- | -------------------------- | --------------------------- | ---------------- | | `openai/gpt-5.5` | omitted / pro...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ 0.006045883055776358, -0.019490430131554604, -0.13320918381214142, 0.035194866359233856, 0.035599857568740845, -0.04754815995693207, 0.018087787553668022, 0.013158150017261505, 0.005184256471693516, -0.006776147522032261, 0.03397149220108986, -0.06026551499962807, 0.021407276391983032, 0.0...
0.04774
that model; direct OpenAI API-key and Azure refs for it remain suppressed. The legacy Codex model prefix is legacy config repaired by doctor. For the common subscription plus native runtime setup, sign in with Codex auth but keep the model ref as `openai/gpt-5.5`. New config should put OpenAI agent auth order under `au...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.06510244309902191, 0.005518092308193445, -0.0544615164399147, 0.025920463725924492, 0.05629013106226921, -0.03789031133055687, -0.05416217818856239, -0.035522207617759705, 0.017843978479504585, 0.018659137189388275, 0.03451727330684662, 0.0427064374089241, 0.04651472717523575, 0.0715727...
-0.00556
### Catalog recovery OpenClaw uses upstream Codex catalog metadata for `gpt-5.5` when it is present. If live Codex discovery omits the `gpt-5.5` row while the account is authenticated, OpenClaw synthesizes that OAuth model row so cron, sub-agent, and configured default-model runs do not fail with `Unknown model`. ## Na...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.06972374767065048, -0.027942650020122528, -0.09437990933656693, -0.0011185940820723772, 0.043254878371953964, -0.11091617494821548, -0.05595162510871887, 0.02454119361937046, 0.010976171121001244, -0.06303586810827255, 0.020605437457561493, -0.04377302527427673, -0.007951396517455578, 0...
0.003121
transparent background" \ --json ``` Use the same `--output-format` and `--background` flags with `openclaw infer image edit` when starting from an input file. `--openai-background` remains available as an OpenAI-specific alias. For ChatGPT/Codex OAuth installs, keep the same `openai/gpt-image-2` ref. When an `openai` ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.01811407506465912, 0.005990718957036734, -0.06465370953083038, 0.06184999272227287, 0.1004636362195015, -0.1067899838089943, -0.005721583031117916, 0.03882985934615135, 0.07189717888832092, 0.01720600575208664, 0.03261294588446617, 0.021404404193162918, 0.06246628239750862, 0.0829365253...
0.030102
a compatibility fallback when the shared `agents.defaults.promptOverlays.gpt5.personality` setting is not set. ## Voice and speech The bundled `openai` plugin registers speech synthesis for the `messages.tts` surface. | Setting | Config path | Default | |---------|------------|---------| | Model | `messages.tts.provide...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.03349044546484947, -0.08827579766511917, -0.026832016184926033, -0.015207321383059025, 0.011847887188196182, -0.011316869407892227, -0.052836041897535324, -0.025797119364142418, -0.04992639645934105, -0.07704158872365952, 0.046463433653116226, -0.02393226884305477, -0.058338649570941925, ...
0.059137
`alloy`, `ash`, `ballad`, `coral`, `echo`, `sage`, `shimmer`, `verse`, `marin`, `cedar`. OpenAI recommends `marin` and `cedar` for the best Realtime quality. This is a separate set from the Text-to-speech voices above; do not assume a TTS voice such as `fable`, `nova`, or `onyx` is valid for Realtime sessions. Backend ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.10492395609617233, -0.06245577335357666, -0.08109107613563538, 0.02108214981853962, -0.02125791646540165, -0.04103695973753929, 0.04908743500709534, -0.08784705400466919, 0.03878850117325783, 0.008366353809833527, -0.0030355574563145638, -0.036091480404138565, -0.011491705663502216, 0.0...
0.128681
in the Azure portal, not the public OpenAI model id. If you create a deployment called `gpt-image-2-prod` that serves `gpt-image-2`: ``` /tool image\_generate model=openai/gpt-image-2-prod prompt="A clean poster" size=1024x1024 count=1 ``` The same deployment-name rule applies to image-generation calls routed through t...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ 0.05599362403154373, -0.027669070288538933, -0.03141177445650101, 0.05303052440285683, 0.12529714405536652, -0.05797746032476425, -0.011355604976415634, -0.053556445986032486, 0.030198900029063225, 0.024710800498723984, 0.05828826501965523, -0.007140107452869415, 0.07815852016210556, 0.055...
-0.002105
when unavailable) This applies to the built-in OpenClaw runtime path and to OpenAI provider hooks used by embedded runs. The native Codex app-server harness manages its own context through Codex and is configured by OpenAI's default agent route or provider/model runtime policy. Useful for compatible endpoints like Azur...
https://github.com/openclaw/openclaw/blob/main/docs/providers/openai.md
main
opebclaw
[ -0.048753462731838226, -0.006732449401170015, -0.05606609582901001, 0.09725888073444366, 0.05527949705719948, -0.07402010262012482, -0.04036283493041992, -0.048916928470134735, 0.029229776933789253, 0.0011965739540755749, 0.04443984106183052, -0.013363978825509548, -0.022781306877732277, 0...
0.094743
Xiaomi MiMo is the API platform for \*\*MiMo\*\* models. OpenClaw includes a bundled Xiaomi plugin with two text-provider presets: - `xiaomi` for pay-as-you-go keys (`sk-...`) - `xiaomi-token-plan` for Token Plan keys (`tp-...`) with regional endpoint presets The same plugin also registers the `xiaomi` speech (TTS) pro...
https://github.com/openclaw/openclaw/blob/main/docs/providers/xiaomi.md
main
opebclaw
[ -0.059214066714048386, -0.08275226503610611, -0.022916393354535103, -0.009579108096659184, -0.03841382637619972, 0.017296716570854187, 0.035806212574243546, 0.03961031138896942, 0.007007735315710306, -0.06109672412276268, 0.11249454319477081, -0.014357536099851131, 0.0273867379873991, 0.01...
0.228869
}, }, } ``` Supported built-in voices include `mimo\_default`, `default\_zh`, `default\_en`, `Mia`, `Chloe`, `Milo`, and `Dean`. Preset-voice models use `audio.voice`, so OpenClaw sends `speakerVoice` for `mimo-v2.5-tts` and `mimo-v2-tts`. Xiaomi's voicedesign model, `mimo-v2.5-tts-voicedesign`, generates the voice fro...
https://github.com/openclaw/openclaw/blob/main/docs/providers/xiaomi.md
main
opebclaw
[ -0.09762829542160034, -0.02527615986764431, 0.02421201765537262, 0.021232999861240387, -0.008562341332435608, -0.08271336555480957, -0.06393559277057648, -0.04599262773990631, 0.01698055863380432, -0.08999806642532349, 0.021447166800498962, -0.04705292358994484, -0.001295951777137816, 0.04...
0.088386
reference. Xiaomi MiMo dashboard and API key management.
https://github.com/openclaw/openclaw/blob/main/docs/providers/xiaomi.md
main
opebclaw
[ -0.0687895342707634, -0.07351750880479813, -0.032544780522584915, -0.04825306683778763, 0.029737044125795364, -0.050465937703847885, 0.042811546474695206, 0.05091705918312073, -0.04076395183801651, 0.005243015941232443, 0.0973040759563446, 0.05714669078588486, 0.13401375710964203, 0.025452...
0.234377
Venice AI provides \*\*privacy-focused AI inference\*\* with support for uncensored models and access to major proprietary models through their anonymized proxy. All inference is private by default — no training on your data, no logging. ## Why Venice in OpenClaw - \*\*Private inference\*\* for open-source models (no l...
https://github.com/openclaw/openclaw/blob/main/docs/providers/venice.md
main
opebclaw
[ -0.05284372344613075, -0.055542971938848495, -0.04009705036878586, 0.05261913686990738, 0.04500145465135574, -0.03692470118403435, -0.0422373004257679, -0.02947145514190197, 0.017578810453414917, -0.05547608435153961, 0.005198202095925808, -0.02541855163872242, -0.04268747568130493, 0.0002...
0.194438
but no Venice tool support | | \*\*Uncensored\*\* | `venice-uncensored` | No content restrictions | ## DeepSeek V4 replay behavior If Venice exposes DeepSeek V4 models such as `venice/deepseek-v4-pro` or `venice/deepseek-v4-flash`, OpenClaw fills the required DeepSeek V4 `reasoning\_content` replay placeholder on assis...
https://github.com/openclaw/openclaw/blob/main/docs/providers/venice.md
main
opebclaw
[ -0.057280175387859344, -0.055016934871673584, 0.016484037041664124, 0.02831588126718998, -0.05683847516775131, -0.0651027038693428, -0.11957110464572906, -0.0575006902217865, -0.0205331202596426, -0.08499983698129654, -0.02392885461449623, -0.03241586685180664, -0.022640621289610863, -0.01...
0.08492
Venice) | 256k | Reasoning, vision | | `grok-41-fast` | Grok 4.1 Fast (via Venice) | 1M | Reasoning, vision | ## Model discovery OpenClaw ships a manifest-backed Venice seed catalog for read-only model listing. Runtime refresh can still discover models from the Venice API, and falls back to the manifest catalog if the ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/venice.md
main
opebclaw
[ -0.04706920310854912, -0.09130140393972397, -0.04266062006354332, 0.022774219512939453, 0.028214527294039726, -0.030215006321668625, -0.07826041430234909, 0.013484392315149307, -0.04418061673641205, -0.050683435052633286, 0.02411242388188839, -0.06559973955154419, -0.006665546912699938, 0....
0.162926
OpenClaw can use \*\*Amazon Bedrock\*\* models via its \*\*Bedrock Converse\*\* streaming provider. Bedrock auth uses the \*\*AWS SDK default credential chain\*\*, not an API key. | Property | Value | | -------- | ----------------------------------------------------------- | | Provider | `amazon-bedrock` | | API | `bed...
https://github.com/openclaw/openclaw/blob/main/docs/providers/bedrock.md
main
opebclaw
[ -0.0017621509032323956, -0.009172485210001469, -0.0926341786980629, -0.004761686082929373, 0.048338521271944046, -0.07032029330730438, -0.03871528059244156, -0.013020160607993603, 0.022036854177713394, 0.012447597458958626, -0.013605485670268536, -0.08516795933246613, 0.04874282330274582, ...
-0.051875
| Default | Description | | ------ | ------- | ----------- | | `enabled` | auto | In auto mode, OpenClaw only enables the implicit Bedrock provider when it sees a supported AWS env marker. Set `true` to force discovery. | | `region` | `AWS\_REGION` / `AWS\_DEFAULT\_REGION` / `us-east-1` | AWS region used for discovery ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/bedrock.md
main
opebclaw
[ -0.013158380053937435, -0.022897690534591675, -0.04685147479176521, 0.05617128685116768, 0.09783223271369934, -0.08937574923038483, -0.04587218537926674, -0.033714789897203445, -0.02044563554227352, 0.028578659519553185, 0.05252452194690704, -0.04823355749249458, 0.001325690420344472, 0.01...
-0.014364
this error, check whether the model supports the requested tier. Bedrock rejects the `temperature` parameter for Claude Opus 4.7. OpenClaw omits `temperature` automatically for any Opus 4.7 Bedrock ref, including foundation model ids, named inference profiles, application inference profiles whose underlying model resol...
https://github.com/openclaw/openclaw/blob/main/docs/providers/bedrock.md
main
opebclaw
[ 0.0031334287486970425, -0.004601442255079746, -0.024080073460936546, 0.037084195762872696, 0.06265594810247421, -0.0010389332892373204, -0.03408518061041832, 0.005332529079169035, -0.016896840184926987, 0.017200903967022896, -0.031244337558746338, -0.12027353793382645, 0.0037063201889395714,...
-0.030806
[inferrs](https://github.com/ericcurtin/inferrs) can serve local models behind an OpenAI-compatible `/v1` API. OpenClaw works with `inferrs` through the generic `openai-completions` path. | Property | Value | | ------------------ | ------------------------------------------------------------------ | | Provider id | `in...
https://github.com/openclaw/openclaw/blob/main/docs/providers/inferrs.md
main
opebclaw
[ -0.06095315143465996, -0.04122675582766533, -0.06663551926612854, 0.05017070099711418, 0.09717416763305664, -0.0458238385617733, -0.07497262954711914, -0.0583094097673893, 0.02244609408080578, -0.012640189379453659, 0.03603918477892876, -0.08136823028326035, 0.005673267412930727, 0.0781965...
0.036407
2 + 2?"}],"stream":false}' ``` ```bash openclaw infer model run \ --model inferrs/google/gemma-4-E2B-it \ --prompt "What is 2 + 2? Reply with one short sentence." \ --json ``` If the first command works but the second fails, check the troubleshooting section below. `inferrs` is treated as a proxy-style OpenAI-compatibl...
https://github.com/openclaw/openclaw/blob/main/docs/providers/inferrs.md
main
opebclaw
[ -0.04322655871510506, -0.03966958075761795, -0.029933933168649673, 0.008749864064157009, 0.02637571096420288, -0.11905885487794876, -0.050060614943504333, -0.009342316538095474, -0.03824308142066002, -0.012760118581354618, 0.04186257719993591, -0.13911305367946625, -0.005940156523138285, 0...
0.111854
OpenClaw integrates with Ollama's native API (`/api/chat`) for hosted cloud models and local/self-hosted Ollama servers. You can use Ollama in three modes: `Cloud + Local` through a reachable Ollama host, `Cloud only` against `https://ollama.com`, or `Local only` against a reachable Ollama host. OpenClaw also registers...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.05298680067062378, 0.011760932393372059, -0.04648159444332123, 0.0891883373260498, -0.02342926524579525, -0.12205673009157181, -0.07653852552175522, -0.0447816364467144, 0.03189286217093468, -0.013281206600368023, 0.03536388278007507, -0.014658183790743351, -0.031678199768066406, -0.032...
0.018742
mode ```bash openclaw onboard --non-interactive \ --auth-choice ollama \ --accept-risk ``` Optionally specify a custom base URL or model: ```bash openclaw onboard --non-interactive \ --auth-choice ollama \ --custom-base-url "http://ollama-host:11434" \ --custom-model-id "qwen3.5:27b" \ --accept-risk ``` \*\*Best for:\*...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.02504587545990944, -0.009118986316025257, -0.08644628524780273, 0.012776646763086319, -0.015742363408207893, -0.08816418796777725, -0.017895672470331192, -0.01707349345088005, 0.051878299564123154, 0.037443727254867554, 0.04051259532570839, -0.040543071925640106, 0.02299250289797783, -0...
0.007489
keeping the catalog aligned with the local Ollama instance. You can use a full ref such as `ollama/:latest` in local `infer model run`; OpenClaw resolves that installed model from Ollama's live catalog without requiring a hand-written `models.json` entry. For signed-in Ollama hosts, some `:cloud` models may be usable t...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.045582856982946396, -0.025780554860830307, -0.03768996149301529, 0.04035087302327156, -0.007619072217494249, -0.07716277241706848, -0.11027345806360245, -0.0016365529736503959, 0.03812291473150253, -0.030273716896772385, 0.06462845951318741, -0.03469890356063843, -0.017558779567480087, ...
-0.001257
a custom remote provider such as `models.providers.ollama-cloud` with `api: "ollama"`, auto-discovery is skipped and you must define models manually. Loopback custom providers such as `http://127.0.0.2:11434` are still treated as local. See the explicit config section below. ## Vision and image description The bundled ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.0387832447886467, 0.0018818448297679424, -0.0023392539005726576, 0.04253794625401497, 0.023335549980401993, -0.08932764828205109, -0.0752621591091156, 0.009853208437561989, 0.04240452125668526, -0.0006498031434603035, 0.06529480963945389, -0.040942732244729996, -0.022751901298761368, 0....
-0.020675
name: "kimi-k2.5:cloud", reasoning: false, input: ["text", "image"], cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, contextWindow: 128000, maxTokens: 8192 } ] } } } } ``` If Ollama is running on a different host or port (explicit config disables auto-discovery, so define models manually): ```json5 { models...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.06536135077476501, 0.05045577883720398, 0.0024432295467704535, 0.05801171436905861, 0.005636121146380901, -0.10188599675893784, -0.07925111055374146, -0.03444211930036545, 0.05688180401921272, -0.01881144940853119, 0.010503104887902737, -0.045252446085214615, -0.056063923984766006, 0.04...
0.087106
}], }, }, }, agents: { defaults: { model: { primary: "ollama-fast/gemma4", fallbacks: ["ollama-large/qwen3.5:27b"], }, }, }, } ``` When OpenClaw sends the request, the active provider prefix is stripped so `ollama-large/qwen3.5:27b` reaches Ollama as `qwen3.5:27b`. Some local models can answer simple prompts but strugg...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.05626680329442024, 0.02814345248043537, 0.02272770367562771, 0.05945136770606041, 0.010910173878073692, -0.13102374970912933, -0.06660228967666626, -0.04859934747219086, -0.006110797636210918, -0.01991376094520092, 0.06553875654935837, -0.07950041443109512, -0.0418180376291275, 0.038207...
0.038163
"https://ollama.com", apiKey: "OLLAMA\_API\_KEY", api: "ollama", models: [{ id: "kimi-k2.5:cloud", name: "kimi-k2.5:cloud", input: ["text"] }], }, }, }, tools: { web: { search: { provider: "ollama" }, }, }, } ``` For a signed-in local daemon, OpenClaw uses the daemon's `/api/experimental/web\_search` proxy. For `https:...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.09060840308666229, 0.03277808427810669, -0.05917363613843918, 0.0536598265171051, -0.01997392810881138, -0.11983928829431534, -0.07559831440448761, -0.04745449125766754, 0.04076011851429939, -0.04493312910199165, 0.02834336645901203, -0.011302386410534382, 0.012021590024232864, -0.02587...
0.094232
ollama/gemma4 --thinking off openclaw agent --model ollama/gemma4 --thinking low ``` You can also set a model default: ```json5 { agents: { defaults: { models: { "ollama/gemma4": { thinking: "low", }, }, }, }, } ``` Per-model `params.think` or `params.thinking` can disable or force Ollama API thinking for a specific co...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.03150520101189613, -0.01590004935860634, -0.02852620929479599, 0.07581889629364014, -0.04542171582579613, -0.12015446275472641, -0.10683488100767136, 0.020446868613362312, -0.030533313751220703, -0.039802487939596176, 0.017876101657748222, -0.029323246330022812, -0.03635256737470627, 0....
0.135589
a GPU-backed model during WSL2 boot, Ollama can pin host memory while the model loads. Hyper-V memory reclaim cannot always reclaim those pinned pages, so Windows can terminate the WSL2 VM, systemd starts Ollama again, and the loop repeats. Common evidence: - repeated WSL2 reboots or terminations from the Windows side ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ 0.005103419069200754, 0.03071664460003376, -0.07111573964357376, 0.09901797771453857, 0.020144950598478317, -0.08332055062055588, -0.04524919018149376, 0.018591493368148804, 0.007229454815387726, -0.07233615219593048, 0.02972436137497425, 0.07153582572937012, -0.05443590134382248, 0.018563...
0.008957
contexts that are larger than your hardware can run comfortably. Native Ollama uses Ollama's own runtime context default unless you set `params.num\_ctx`. Cap both OpenClaw's budget and Ollama's request context when you want predictable first-token latency: ```json5 { models: { providers: { ollama: { contextWindow: 327...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama.md
main
opebclaw
[ -0.07250135391950607, 0.013001133687794209, -0.01759517937898636, 0.053776971995830536, -0.03343331068754196, -0.17087222635746002, -0.028785748407244682, 0.025764258578419685, 0.027164120227098465, -0.04750077798962593, -0.00870233029127121, -0.06818080693483353, -0.04833727329969406, -0....
0.100451
OpenClaw ships a bundled `fal` provider for hosted image, video, and music generation. | Property | Value | | -------- | ------------------------------------------------------------- | | Provider | `fal` | | Auth | `FAL\_KEY` (canonical; `FAL\_API\_KEY` also works as a fallback) | | API | fal model endpoints | ## Getti...
https://github.com/openclaw/openclaw/blob/main/docs/providers/fal.md
main
opebclaw
[ -0.058328669518232346, -0.009869815781712532, -0.0671970471739769, 0.019236883148550987, 0.09350954741239548, -0.050739921629428864, -0.0472397543489933, 0.018744582310318947, 0.010945620015263557, -0.03371815010905266, 0.045666199177503586, -0.0651085302233696, 0.028570881113409996, 0.048...
-0.016651
}, }, } ``` ## Music generation The bundled `fal` plugin also registers a music-generation provider for the shared `music\_generate` tool. | Capability | Value | | ------------- | ------------------------------------------------------------------------------------------------------ | | Default model | `fal/fal-ai/minim...
https://github.com/openclaw/openclaw/blob/main/docs/providers/fal.md
main
opebclaw
[ -0.06598009914159775, -0.07140855491161346, -0.1366593837738037, -0.019568730145692825, 0.023887844756245613, 0.06558454781770706, -0.05047765001654625, 0.01786952093243599, 0.006430062465369701, -0.0542026124894619, 0.06466872245073318, -0.07535538822412491, -0.04393579438328743, -0.07594...
0.080388
SGLang serves open-weight models via an OpenAI-compatible HTTP API. OpenClaw connects to SGLang using the `openai-completions` provider family with auto-discovery of available models. | Property | Value | | ------------------------- | ------------------------------------------------------------ | | Provider id | `sglan...
https://github.com/openclaw/openclaw/blob/main/docs/providers/sglang.md
main
opebclaw
[ -0.08947816491127014, -0.05215185508131981, -0.05909309163689613, 0.06407064199447632, 0.0004633383941836655, -0.03591107204556465, -0.016601882874965668, -0.01149008795619011, -0.0024337749928236008, -0.05713973566889763, 0.07595551759004593, -0.0830162838101387, 0.017624668776988983, 0.0...
0.100343
Azure Speech is an Azure AI Speech text-to-speech provider. In OpenClaw it synthesizes outbound reply audio as MP3 by default, native Ogg/Opus for voice notes, and 8 kHz mulaw audio for telephony channels such as Voice Call. OpenClaw uses the Azure Speech REST API directly with SSML and sends the provider-owned output ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/azure-speech.md
main
opebclaw
[ -0.05957295000553131, -0.02385856583714485, -0.09899023175239563, -0.0006201476790010929, -0.00837808009237051, 0.027868935838341713, -0.010809545405209064, -0.08359798043966293, 0.05549989268183708, 0.008687673136591911, -0.01822621002793312, -0.020054327324032784, -0.012382891960442066, ...
0.154353
[Chutes](https://chutes.ai) exposes open-source model catalogs through an OpenAI-compatible API. OpenClaw supports both browser OAuth and direct API-key auth for the bundled `chutes` provider. | Property | Value | | -------- | ---------------------------- | | Provider | `chutes` | | API | OpenAI-compatible | | Base URL...
https://github.com/openclaw/openclaw/blob/main/docs/providers/chutes.md
main
opebclaw
[ -0.08220367878675461, -0.06775844842195511, -0.125078022480011, 0.010038125328719616, 0.006346145179122686, -0.1074303388595581, 0.001672757207415998, 0.01143564935773611, 0.036104850471019745, -0.046548619866371155, 0.07411698251962662, -0.020617758855223656, 0.04551278054714203, 0.020884...
0.090299
OpenClaw ships a bundled `alibaba` plugin that registers a video-generation provider for Wan models on Alibaba Model Studio (the international name for DashScope). The plugin is enabled by default; you only need to set an API key. | Property | Value | | ---------------- | -----------------------------------------------...
https://github.com/openclaw/openclaw/blob/main/docs/providers/alibaba.md
main
opebclaw
[ -0.07511746138334274, -0.06800980865955353, -0.1036606878042221, 0.03188781067728996, -0.028225718066096306, 0.008846473880112171, -0.03798709809780121, 0.011280962266027927, 0.027510028332471848, 0.034864120185375214, 0.03132956475019455, -0.09138209372758865, 0.007418201770633459, 0.0326...
-0.017662
rotation, cooldown, and override mechanics. Both bundled plugins talk to DashScope and accept overlapping API keys. Use: - `alibaba/wan\*.\*` ids to drive the dedicated Wan video provider documented on this page. - `qwen/\*` ids for Qwen chat, embedding, and media understanding (see [Qwen](/providers/qwen)). Setting `M...
https://github.com/openclaw/openclaw/blob/main/docs/providers/alibaba.md
main
opebclaw
[ -0.07694245129823685, -0.11205679178237915, -0.037487272173166275, -0.0651589035987854, -0.0483151450753212, -0.00939270295202732, 0.0042422739788889885, 0.024588655680418015, 0.01826632022857666, 0.050921060144901276, 0.06518766283988953, -0.0028167671989649534, 0.012355844490230083, 0.09...
-0.050254
OpenClaw includes a bundled StepFun provider plugin with two provider ids: - `stepfun` for the standard endpoint - `stepfun-plan` for the Step Plan endpoint Standard and Step Plan are \*\*separate providers\*\* with different endpoints and model ref prefixes (`stepfun/...` vs `stepfun-plan/...`). Use a China key with t...
https://github.com/openclaw/openclaw/blob/main/docs/providers/stepfun.md
main
opebclaw
[ -0.08478488773107529, 0.02727500908076763, -0.03128480166196823, 0.04082107916474342, 0.021573340520262718, -0.04734289273619652, -0.10264991968870163, -0.016788510605692863, 0.031016908586025238, -0.05271744355559349, 0.036692049354314804, -0.04477054998278618, -0.020281290635466576, -0.0...
-0.002384
so both surfaces can be discovered together. - Use `openclaw models list` and `openclaw models set ` to inspect or switch models. For the broader provider overview, see [Model providers](/concepts/model-providers). ## Related Overview of all providers, model refs, and failover behavior. Full config schema for providers...
https://github.com/openclaw/openclaw/blob/main/docs/providers/stepfun.md
main
opebclaw
[ -0.09156861156225204, -0.07956267148256302, -0.027898034080863, 0.043670788407325745, 0.043203551322221756, -0.07212889194488525, -0.07848737388849258, 0.036775678396224976, -0.02965068817138672, -0.02428348734974861, 0.05767277255654335, -0.025402948260307312, -0.007797446101903915, 0.005...
0.073835
[Together AI](https://together.ai) provides access to leading open-source models including Llama, DeepSeek, Kimi, and more through a unified API. | Property | Value | | -------- | ----------------------------- | | Provider | `together` | | Auth | `TOGETHER\_API\_KEY` | | API | OpenAI-compatible | | Base URL | `https://...
https://github.com/openclaw/openclaw/blob/main/docs/providers/together.md
main
opebclaw
[ -0.05608943849802017, -0.020356079563498497, -0.10292080044746399, 0.0215291827917099, 0.010831906460225582, -0.04388470947742462, -0.09184858947992325, 0.006033009383827448, 0.016985828056931496, -0.02694687992334366, -0.00012138298188801855, -0.04877108708024025, 0.019256118685007095, 0....
0.093477
OpenClaw uses ElevenLabs for text-to-speech, batch speech-to-text with Scribe v2, and streaming STT with Scribe v2 Realtime. | Capability | OpenClaw surface | Default | | ------------------------ | -------------------------------------------------------------------- | ------------------------ | | Text-to-speech | `mess...
https://github.com/openclaw/openclaw/blob/main/docs/providers/elevenlabs.md
main
opebclaw
[ -0.08747593313455582, -0.08563397079706192, -0.07662852853536606, -0.04321836680173874, -0.026102393865585327, -0.04483584314584732, -0.01223229244351387, -0.028774352744221687, -0.01789945363998413, -0.07409857958555222, 0.03793380782008171, -0.07600446790456772, -0.054299499839544296, -0...
0.094707
Ollama Cloud is Ollama's hosted model API. It lets OpenClaw call Ollama-hosted models directly, without installing a local Ollama server or signing a local Ollama app into cloud mode. Use provider id `ollama-cloud` and model refs like `ollama-cloud/kimi-k2.6`. This page is for direct cloud-only routing. The provider us...
https://github.com/openclaw/openclaw/blob/main/docs/providers/ollama-cloud.md
main
opebclaw
[ -0.02249850705265999, -0.024155467748641968, -0.02025563269853592, 0.09159217774868011, 0.0006620609783567488, -0.10842011123895645, -0.08729984611272812, -0.054739806801080704, 0.03628776594996452, -0.008384806104004383, 0.027780968695878983, -0.020395241677761078, -0.03539925068616867, -...
-0.002565
OpenClaw provides `pixverse` as an official external plugin for hosted PixVerse video generation. The plugin registers the `pixverse` provider against the `videoGenerationProviders` contract. | Property | Value | | ------------------ | -------------------------------------------------------------------- | | Provider id...
https://github.com/openclaw/openclaw/blob/main/docs/providers/pixverse.md
main
opebclaw
[ -0.04502429813146591, -0.07871873676776886, -0.06658155471086502, -0.02769472636282444, 0.009185527451336384, -0.0035766970831900835, -0.10340948402881622, 0.03818190097808838, 0.03335341811180115, 0.009832571260631084, 0.03493313118815422, -0.07835406064987183, 0.05557398125529289, -0.007...
-0.01053
tool parameters, provider selection, and async behavior. Agent default settings including video generation model.
https://github.com/openclaw/openclaw/blob/main/docs/providers/pixverse.md
main
opebclaw
[ -0.004772948566824198, -0.08824525773525238, -0.11764246225357056, -0.006436917930841446, 0.013104338198900223, 0.029499324038624763, -0.06839807331562042, -0.03940831497311592, 0.025588320568203926, -0.02395588532090187, 0.0593741275370121, -0.019992806017398834, -0.05326906964182854, -0....
0.090148
vLLM can serve open-source (and some custom) models via an \*\*OpenAI-compatible\*\* HTTP API. OpenClaw connects to vLLM using the `openai-completions` API. OpenClaw can also \*\*auto-discover\*\* available models from vLLM when you opt in with `VLLM\_API\_KEY` (any value works if your server does not enforce auth). Us...
https://github.com/openclaw/openclaw/blob/main/docs/providers/vllm.md
main
opebclaw
[ -0.029371613636612892, -0.004781231749802828, -0.11467494070529938, 0.09302212297916412, -0.0191294364631176, -0.04246598109602928, -0.03634743392467499, -0.0033938486594706774, 0.0471639409661293, -0.03024822100996971, 0.007713760249316692, -0.08341142535209656, -0.011172554455697536, 0.0...
0.116212
``` OpenClaw maps `/think off` to: ```json { "chat\_template\_kwargs": { "enable\_thinking": false, "preserve\_thinking": true } } ``` Non-`off` thinking levels send `enable\_thinking: true`. If your endpoint expects DashScope-style top-level flags instead, use `compat.thinkingFormat: "qwen"` to send `enable\_thinking`...
https://github.com/openclaw/openclaw/blob/main/docs/providers/vllm.md
main
opebclaw
[ -0.04202570766210556, 0.026992259547114372, -0.007900578901171684, 0.04862310364842415, -0.03567489609122276, -0.07240326702594757, -0.015826132148504257, 0.011367307975888252, 0.06349648535251617, -0.02024427428841591, 0.028569040820002556, -0.09864537417888641, -0.06171153858304024, 0.06...
0.10519
out of exact-origin trust. If requests fail with auth errors, set a real `VLLM\_API\_KEY` that matches your server configuration, or configure the provider explicitly under `models.providers.vllm`. If your vLLM server does not enforce auth, any non-empty value for `VLLM\_API\_KEY` works as an opt-in signal for OpenClaw...
https://github.com/openclaw/openclaw/blob/main/docs/providers/vllm.md
main
opebclaw
[ -0.07346569001674652, -0.02455908991396427, -0.06872107088565826, 0.04759802669286728, -0.06517506390810013, -0.07843843847513199, -0.08002404868602753, -0.046268902719020844, 0.020363222807645798, -0.018526369705796242, 0.05253952741622925, -0.11195995658636093, 0.010398587211966515, 0.04...
-0.031261
OpenClaw includes a bundled Mistral plugin that registers four contracts: chat completions, media understanding (Voxtral batch transcription), realtime STT for Voice Call (Voxtral Realtime), and memory embeddings (`mistral-embed`). | Property | Value | | ---------------- | ------------------------------------------- | ...
https://github.com/openclaw/openclaw/blob/main/docs/providers/mistral.md
main
opebclaw
[ -0.03734612092375755, -0.0579197034239769, -0.08522409945726395, 0.017118200659751892, 0.03936256095767021, -0.0017401198856532574, -0.014486541971564293, 0.008183414116501808, 0.04558352008461952, -0.05740322545170784, 0.0468151718378067, -0.029042275622487068, -0.01616453006863594, -0.00...
0.084489
apiKey: "${MISTRAL\_API\_KEY}", targetStreamingDelayMs: 800, }, }, }, }, }, }, }, } ``` OpenClaw defaults Mistral realtime STT to `pcm\_mulaw` at 8 kHz so Voice Call can forward Twilio media frames directly. Use `encoding: "pcm\_s16le"` and a matching `sampleRate` only if your upstream stream is already raw PCM. ## Adv...
https://github.com/openclaw/openclaw/blob/main/docs/providers/mistral.md
main
opebclaw
[ 0.006909554824233055, -0.038352303206920624, -0.003513897303491831, -0.07504849135875702, 0.010562291368842125, -0.08276913315057755, 0.05393941327929497, 0.04396187886595726, 0.042987458407878876, -0.02180435135960579, -0.03719666972756386, -0.07212533056735992, -0.04269464313983917, 0.02...
0.059987
Deepgram is a speech-to-text API. In OpenClaw it is used for inbound audio/voice-note transcription through `tools.media.audio` and for Voice Call streaming STT through `plugins.entries.voice-call.config.streaming`. For batch transcription, OpenClaw uploads the complete audio file to Deepgram and injects the transcript...
https://github.com/openclaw/openclaw/blob/main/docs/providers/deepgram.md
main
opebclaw
[ -0.125576451420784, -0.04917182773351669, -0.06644878536462784, -0.039269790053367615, 0.005138188134878874, -0.0737205520272255, -0.11404303461313248, 0.008289284072816372, 0.00007261415885295719, -0.09835519641637802, -0.044469624757766724, 0.012318387627601624, -0.031717121601104736, -0...
0.16987
Cloudflare AI Gateway sits in front of provider APIs and lets you add analytics, caching, and controls. For Anthropic, OpenClaw uses the Anthropic Messages API through your Gateway endpoint. | Property | Value | | ------------- | ---------------------------------------------------------------------------------------- |...
https://github.com/openclaw/openclaw/blob/main/docs/providers/cloudflare-ai-gateway.md
main
opebclaw
[ -0.006128388457000256, -0.06575377285480499, -0.1210700124502182, 0.025808105245232582, 0.0352511890232563, -0.0354011170566082, 0.030616488307714462, -0.027875710278749466, 0.10899729281663895, 0.05910316854715347, 0.015906650573015213, -0.015627264976501465, 0.0025849889498203993, -0.061...
0.052383
The bundled Vydra plugin adds: - Image generation via `vydra/grok-imagine` - Video generation via `vydra/veo3` and `vydra/kling` - Speech synthesis via Vydra's ElevenLabs-backed TTS route OpenClaw uses the same `VYDRA\_API\_KEY` for all three capabilities. | Property | Value | | --------------- | ----------------------...
https://github.com/openclaw/openclaw/blob/main/docs/providers/vydra.md
main
opebclaw
[ -0.05529116839170456, -0.006113007199019194, -0.1288866549730301, -0.06567816436290741, 0.016011781990528107, -0.04124954342842102, -0.09778179973363876, -0.011432782746851444, -0.017579378560185432, -0.04083646833896637, 0.075398288667202, -0.0730142891407013, 0.022444866597652435, -0.027...
0.031057
LM Studio is a friendly yet powerful app for running open-weight models on your own hardware. It lets you run llama.cpp (GGUF) or MLX models (Apple Silicon). Comes in a GUI package or headless daemon (`llmster`). For product and setup docs, see [lmstudio.ai](https://lmstudio.ai/). ## Quick start 1. Install LM Studio (d...
https://github.com/openclaw/openclaw/blob/main/docs/providers/lmstudio.md
main
opebclaw
[ 0.045732561498880386, -0.11343469470739365, 0.012394484132528305, -0.008562915027141571, -0.011457273736596107, -0.01919923536479473, -0.1402096450328827, 0.09940701723098755, -0.002786266850307584, 0.016507769003510475, 0.025144390761852264, -0.08102870732545853, 0.036254048347473145, -0....
0.041113
that contains `off`/`on` reasoning maps is normalized the same way when the catalog is loaded. ### Explicit configuration ```json5 { models: { providers: { lmstudio: { baseUrl: "http://localhost:1234/v1", apiKey: "${LM\_API\_TOKEN}", api: "openai-completions", models: [ { id: "qwen/qwen3-coder-next", name: "Qwen 3 Code...
https://github.com/openclaw/openclaw/blob/main/docs/providers/lmstudio.md
main
opebclaw
[ 0.010017748922109604, -0.054158806800842285, 0.007976274006068707, 0.011143818497657776, 0.08037259429693222, -0.004539866466075182, -0.0652351900935173, 0.02704121731221676, -0.008646789938211441, 0.029103059321641922, 0.07707216590642929, -0.06325644254684448, -0.005169355776160955, 0.04...
0.066622
The Volcengine provider gives access to Doubao models and third-party models hosted on Volcano Engine, with separate endpoints for general and coding workloads. The same bundled plugin can also register Volcengine Speech as a TTS provider. | Detail | Value | | ---------- | ----------------------------------------------...
https://github.com/openclaw/openclaw/blob/main/docs/providers/volcengine.md
main
opebclaw
[ -0.05585857480764389, -0.06449056416749954, -0.047182291746139526, -0.07018154114484787, -0.053584255278110504, -0.00030484687886200845, -0.06495807319879532, 0.06737539917230606, 0.002243271330371499, -0.051409874111413956, 0.00924012716859579, -0.08878082036972046, -0.05873289331793785, ...
0.166985
Console applications: ```bash export VOLCENGINE\_TTS\_APPID="speech\_app\_id" export VOLCENGINE\_TTS\_TOKEN="speech\_access\_token" export VOLCENGINE\_TTS\_CLUSTER="volcano\_tts" ``` ## Advanced configuration `openclaw onboard --auth-choice volcengine-api-key` currently sets `volcengine-plan/ark-code-latest` as the def...
https://github.com/openclaw/openclaw/blob/main/docs/providers/volcengine.md
main
opebclaw
[ -0.005072046536952257, -0.037849124521017075, -0.024935070425271988, 0.022135121747851372, -0.0017052023904398084, -0.06603558361530304, -0.06826595216989517, -0.02222595177590847, 0.02817767672240734, -0.049271464347839355, 0.000007672913852729835, -0.08194126188755035, -0.03659551590681076...
-0.001999
OpenClaw can use many LLM providers. Pick a provider, authenticate, then set the default model as `provider/model`. Looking for chat channel docs (WhatsApp/Telegram/Discord/Slack/Mattermost (plugin)/etc.)? See [Channels](/channels). ## Quick start 1. Authenticate with the provider (usually via `openclaw onboard`). 2. S...
https://github.com/openclaw/openclaw/blob/main/docs/providers/index.md
main
opebclaw
[ -0.04878341034054756, -0.07943587750196457, -0.017419399693608284, -0.005825937259942293, -0.03485669195652008, -0.032292187213897705, -0.014581795781850815, -0.031434886157512665, 0.0702935978770256, 0.04374624788761139, -0.014152680523693562, -0.06889075040817261, -0.018303675577044487, ...
0.092777
OpenClaw can use many LLM providers. Pick one, authenticate, then set the default model as `provider/model`. ## Quick start (two steps) 1. Authenticate with the provider (usually via `openclaw onboard`). 2. Set the default model: ```json5 { agents: { defaults: { model: { primary: "anthropic/claude-opus-4-6" } } }, } ``...
https://github.com/openclaw/openclaw/blob/main/docs/providers/models.md
main
opebclaw
[ -0.06140591949224472, -0.05531919002532959, -0.04008578509092331, -0.005517187062650919, 0.008468490093946457, -0.04562046751379967, -0.06605314463376999, -0.020433852449059486, 0.019423946738243103, -0.0026557359378784895, 0.03370463103055954, -0.03709694743156433, -0.01269014086574316, 0...
0.0402