repo
stringclasses
15 values
fix_commit
stringlengths
40
40
buggy_commit
stringlengths
40
40
message
stringlengths
3
64.3k
files
listlengths
1
300
timestamp
timestamp[s]date
2013-03-13 20:45:00
2026-04-11 07:48:46
huggingface/transformers
c23a1c19322152dbc89cfaa5f22310bd0afb92d0
a3f82328eda0a1b37bc49cb66d168d74327b60c0
Add-helium (#35669) * Add the helium model. * Add a missing helium. * And add another missing helium. * Use float for the rmsnorm mul. * Add the Helium tokenizer converter. * Add the pad token as suggested by Arthur. * Update the RMSNorm + some other tweaks. * Fix more rebase issues. * fix copies and style * ...
[ { "path": "docs/source/en/_toctree.yml", "patch": "@@ -452,6 +452,8 @@\n title: Granite\n - local: model_doc/granitemoe\n title: GraniteMoe\n+ - local: model_doc/helium\n+ title: Helium\n - local: model_doc/herbert\n title: HerBERT\n - local: model_doc/i...
2025-01-13T17:41:15
rust-lang/rust
9bb1008e1d8da55be5e383b552e5e9e3774e37fe
3cb9966235a981cfea1bbf2536ebd5021bcbca5e
Avoid empty identifiers for delegate params and args. Instead use `argN`. The empty identifiers could flow to `Liveness::should_warn`, where they would trigger a bounds error. Fixes #139512.
[ { "path": "compiler/rustc_ast_lowering/src/delegation.rs", "patch": "@@ -47,7 +47,7 @@ use rustc_errors::ErrorGuaranteed;\n use rustc_hir::def_id::DefId;\n use rustc_middle::span_bug;\n use rustc_middle::ty::{Asyncness, ResolverAstLowering};\n-use rustc_span::{Ident, Span};\n+use rustc_span::{Ident, Span, S...
2025-04-10T03:44:57
ollama/ollama
6b5db12aa2f8841629cac89d9deb87486512fdb4
612f0a17d32cf7fe9bde5f31703a37b49c98ea2f
mlx: remove stale x86 libmlx library (#15443) Fixes #15433
[ { "path": "scripts/build_darwin.sh", "patch": "@@ -200,10 +200,15 @@ _build_macapp() {\n # Copy .so files from both architectures (names don't collide: arm64=libggml-cpu.so, amd64=libggml-cpu-*.so)\n cp dist/darwin-arm64/lib/ollama/*.so dist/Ollama.app/Contents/Resources/ 2>/dev/null || true...
2026-04-09T03:51:47
vercel/next.js
f3d3153f213f9e953a704048e79d9ff81f1b17b0
a012f64a3c0b8c23661ff85734cbdbeb5c06f409
bugfixes for the error overlay (#134)
[ { "path": "crates/turbo-tasks-memory/src/memory_backend.rs", "patch": "@@ -8,7 +8,7 @@ use std::{\n time::Duration,\n };\n \n-use anyhow::Result;\n+use anyhow::{bail, Result};\n use event_listener::EventListener;\n use flurry::HashMap as FHashMap;\n use tokio::task::futures::TaskLocalFuture;\n@@ -283,6 ...
2022-10-20T16:15:13
nodejs/node
15e136d9c5c31041c6202179f5847fd518e93c23
5eb9f3c91ce48a1c6818aaf4eddb0b8ff8071ecc
test: fix http-agent-destroyed-socket cb not firing A whole part of the test-http-agent-destroyed-socket test was not running as the semantics of http events changed slightly and were no longer triggering the expected event. Instead listen to the same event on the socket to verify that the code being tested is still w...
[ { "path": "test/parallel/test-http-agent-destroyed-socket.js", "patch": "@@ -44,24 +44,11 @@ const server = http.createServer(common.mustCall((req, res) => {\n // assert request2 is queued in the agent\n const key = agent.getName(requestOptions);\n assert.strictEqual(agent.requests[key].length, ...
2018-04-13T10:20:47
golang/go
142a76530cf610fe02d151727fa0d8038c552127
48ddf7012875014d3cab4a02002799a520b087a1
go/types: improve the positioning of broken import errors The heuristic gopls uses to guess error spans can get tripped-up on certain valid characters in an import path (for example '-'). Update the error for broken imports to capture the full import path span, so that gopls doesn't need to rely on heuristics. Chang...
[ { "path": "src/go/types/resolver.go", "patch": "@@ -130,7 +130,7 @@ func (check *Checker) filename(fileNo int) string {\n \treturn fmt.Sprintf(\"file[%d]\", fileNo)\n }\n \n-func (check *Checker) importPackage(pos token.Pos, path, dir string) *Package {\n+func (check *Checker) importPackage(at positioner, p...
2021-03-09T19:22:38
huggingface/transformers
2fa876d2d824123b80ced9d689f75a153731769b
e6f9b0346472509ac08a386de22bab2431edbe3e
[tests] make cuda-only tests device-agnostic (#35607) * intial commit * remove unrelated files * further remove * Update test_trainer.py * fix style
[ { "path": "tests/fsdp/test_fsdp.py", "patch": "@@ -32,7 +32,6 @@\n require_accelerate,\n require_fsdp,\n require_torch_accelerator,\n- require_torch_gpu,\n require_torch_multi_accelerator,\n slow,\n torch_device,\n@@ -288,7 +287,7 @@ def test_training_and_can_resume_normally(self,...
2025-01-13T13:48:39
rust-lang/rust
3cb9966235a981cfea1bbf2536ebd5021bcbca5e
51548ce71fe80b5ca7aef00e6f1bf2491df98c79
Add a HIR pretty printing test for delegation. Note that some of the output is currently bogus, with missing params and args: ``` fn add(: _, : _) -> _ { m::add(, ) } ``` The next commit will fix this.
[ { "path": "tests/pretty/hir-delegation.pp", "patch": "@@ -0,0 +1,23 @@\n+//@ pretty-compare-only\n+//@ pretty-mode:hir\n+//@ pp-exact:hir-delegation.pp\n+\n+#![allow(incomplete_features)]#![feature(fn_delegation)]\n+#[prelude_import]\n+use ::std::prelude::rust_2015::*;\n+#[macro_use]\n+extern crate std;\n+\...
2025-04-10T01:49:50
ollama/ollama
612f0a17d32cf7fe9bde5f31703a37b49c98ea2f
673726fa0eedfc13c4852ff842e80eb1b390351d
fix: improve error message for unknown input item type in responses API (#15424) The default branch in unmarshalResponsesInputItem had two issues: - It referenced typeField.Type instead of itemType; these differ when the shorthand role-based format promotes an empty type to "message", meaning an unhandled type wou...
[ { "path": "openai/responses.go", "patch": "@@ -281,7 +281,10 @@ func unmarshalResponsesInputItem(data []byte) (ResponsesInputItem, error) {\n \t\t}\n \t\treturn reasoning, nil\n \tdefault:\n-\t\treturn nil, fmt.Errorf(\"unknown input item type: %s\", typeField.Type)\n+\t\tif itemType == \"\" {\n+\t\t\tretur...
2026-04-09T00:41:12
nodejs/node
7dd6d864cb3c4e5411bc8d29bfdf54b969126293
daed72f224af5454269362b3d3e3e8d1b80ceb55
doc: fix suspicious heading emphasis in n-api.md PR-URL: https://github.com/nodejs/node/pull/20086 Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com> Reviewed-By: Kyle Farnung <kfarnung@microsoft.com>
[ { "path": "doc/api/n-api.md", "patch": "@@ -3475,7 +3475,7 @@ is sufficient and appropriate. Use of the `napi_make_callback` function\n may be required when implementing custom async behavior that does not use\n `napi_create_async_work`.\n \n-### *napi_open_callback_scope*\n+### napi_open_callback_scope\n <...
2018-04-16T21:23:54
golang/go
18510ae88ffcb9c4a914805fde3e613539f9b6dc
e4f3cfadf618df5135bf0952507ab491975cceb5
runtime, cmd/link/internal/ld: disable memory profiling when data unreachable If runtime.MemProfile is unreachable, default to not collecting any memory profiling samples, to save memory on the hash table. Fixes #42347 Change-Id: I9a4894a5fc77035fe59b1842e1ec77a1182e70c1 Reviewed-on: https://go-review.googlesource.c...
[ { "path": "src/cmd/link/internal/ld/ld_test.go", "patch": "@@ -224,3 +224,103 @@ func testWindowsBuildmodeCSharedASLR(t *testing.T, useASLR bool) {\n \t\tt.Error(\"IMAGE_DLLCHARACTERISTICS_DYNAMIC_BASE flag should not be set\")\n \t}\n }\n+\n+// TestMemProfileCheck tests that cmd/link sets\n+// runtime.disa...
2021-03-08T04:52:48
huggingface/transformers
84a6789145c3d728f2e405d31e9a35df5d74f05c
87089176d9a205d69e7d4e152fc25950d77580d2
Enable different torch dtype in sub models (#34873) * fix * fix test * add tests * add more tests * fix tests * supposed to be a torch.dtype test * handle BC and make fp32 default
[ { "path": "src/transformers/configuration_utils.py", "patch": "@@ -994,8 +994,11 @@ def dict_torch_dtype_to_str(self, d: Dict[str, Any]) -> None:\n converts torch.dtype to a string of just the type. For example, `torch.float32` get converted into *\"float32\"*\n string, which can then be sto...
2025-01-13T12:42:08
ollama/ollama
b5918f9785bd293a51682ae86085e566e5ab8a4c
d17f482d50ed46957ae11b534e3593d9786482c0
pull/push: refine safetensors (#14946) * pull: refine safetensors pull - Body drain in resolve() — drain response body before close so Go's HTTP client can reuse TCP connections instead of opening a new one per blob (1,075 extra TCP+TLS handshakes eliminated) - Skip speed recording for tiny blobs (<100KB) — p...
[ { "path": "x/imagegen/transfer/download.go", "patch": "@@ -117,14 +117,25 @@ func (d *downloader) download(ctx context.Context, blob Blob) error {\n \t\tstart := time.Now()\n \t\tn, err := d.downloadOnce(ctx, blob)\n \t\tif err == nil {\n-\t\t\tif s := time.Since(start).Seconds(); s > 0 {\n-\t\t\t\td.speeds...
2026-04-08T21:15:39
vercel/next.js
4bb347ae397131aa7d2e319ffc10f4817f997289
be1537881ca0893f8a422b2cd6d59ad5ae8a4004
error overlay (vercel/turbo#39) transferred from https://github.com/vercel/turbo-tooling/pull/396 Co-authored-by: Tobias Koppers <tobias.koppers@googlemail.com>
[ { "path": "packages/next-swc/crates/next-core/js/build.mjs", "patch": "@@ -0,0 +1,162 @@\n+/**\n+ * Build npm package to be able to embed them in the binary\n+ */\n+\n+import { mkdir, writeFile, rm, readFile } from \"node:fs/promises\";\n+import { existsSync } from \"node:fs\";\n+import { createRequire } fr...
2022-10-20T16:04:19
huggingface/transformers
b8c34d97fcf6685a480e974f9c88bf04cb3088a9
cd44bdb4b895b9e442476ee556d53ec71f9efe87
Fix whisper compile (#35413) Fix compile error Signed-off-by: jiqing-feng <jiqing.feng@intel.com>
[ { "path": "src/transformers/models/whisper/generation_whisper.py", "patch": "@@ -1571,7 +1571,7 @@ def detect_language(\n )\n \n with torch.no_grad():\n- logits = self(**inputs, decoder_input_ids=decoder_input_ids).logits[:, -1]\n+ logits = self(**inputs, decoder_input_...
2025-01-13T10:31:51
nodejs/node
2e76b175edfecf1d667b2e3bde907f1120f5da49
addc04be0b4a3cc4de336f7bc2c7fdefff249581
doc: fix two sorting nits in fs.md PR-URL: https://github.com/nodejs/node/pull/20078 Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
[ { "path": "doc/api/fs.md", "patch": "@@ -320,6 +320,13 @@ fs.watch('./tmp', { encoding: 'buffer' }, (eventType, filename) => {\n });\n ```\n \n+### Event: 'close'\n+<!-- YAML\n+added: REPLACEME\n+-->\n+\n+Emitted when the watcher stops watching for changes.\n+\n ### Event: 'error'\n <!-- YAML\n added: v0.5....
2018-04-16T15:03:31
golang/go
e4f3cfadf618df5135bf0952507ab491975cceb5
382851c1fd135a99efbe128a3be0ce466d42506f
net: don't append a dot to TXT records on Plan 9 TXT records are not domain names, so no need to call absDomainName. The output now matches the pure Go resolver. Fixes #44619 Change-Id: I1ebf09152ff5c0446d2e2b4c26671358892d9dc9 Reviewed-on: https://go-review.googlesource.com/c/go/+/296589 Reviewed-by: David du Colom...
[ { "path": "src/net/lookup_plan9.go", "patch": "@@ -308,7 +308,7 @@ func (*Resolver) lookupTXT(ctx context.Context, name string) (txt []string, err\n \t}\n \tfor _, line := range lines {\n \t\tif i := bytealg.IndexByteString(line, '\\t'); i >= 0 {\n-\t\t\ttxt = append(txt, absDomainName([]byte(line[i+1:])))\...
2021-02-25T20:16:25
ollama/ollama
f86a969f27ec48cc516bfadc8d566807a14870bd
9fa80a1660bc8ba28a77e6e91cc14d1c2b1cb652
responses: add support for fn call output arrays (#15406) In addition to strings (which we already supported), OpenResponses supports arrays of text content, image content, or file content (see <https://www.openresponses.org/reference#object-FunctionCallOutput-title>). We were missing support for these arrays, which c...
[ { "path": "openai/responses.go", "patch": "@@ -4,13 +4,15 @@ import (\n \t\"encoding/json\"\n \t\"fmt\"\n \t\"math/rand\"\n+\t\"strings\"\n \t\"time\"\n \n \t\"github.com/ollama/ollama/api\"\n )\n \n // ResponsesContent is a discriminated union for input content types.\n-// Concrete types: ResponsesTextCont...
2026-04-07T23:47:30
rust-lang/rust
328b4fa9fe69d9d1ba5fb5329b33eee71e29329b
76d5a26fb2f811a28b30853d68371616c7cdd0f4
Rename some `name` variables as `ident`. It bugs me when variables of type `Ident` are called `name`. It leads to silly things like `name.name`. `Ident` variables should be called `ident`, and `name` should be used for variables of type `Symbol`. This commit improves things by by doing `s/name/ident/` on a bunch of `...
[ { "path": "clippy_lints/src/bool_assert_comparison.rs", "patch": "@@ -53,7 +53,7 @@ fn is_impl_not_trait_with_bool_out<'tcx>(cx: &LateContext<'tcx>, ty: Ty<'tcx>) -\n .not_trait()\n .filter(|trait_id| implements_trait(cx, ty, *trait_id, &[]))\n .and_then(|trait_id| {\n- cx...
2025-04-08T02:23:07
huggingface/transformers
cd44bdb4b895b9e442476ee556d53ec71f9efe87
15bd3e61f8d3680ca472c9314ad07584d20f7b81
Fix device in rope module when using dynamic updates (#35608) fix rope device
[ { "path": "src/transformers/models/aria/modeling_aria.py", "patch": "@@ -754,6 +754,9 @@ def _dynamic_frequency_update(self, position_ids, device):\n self.max_seq_len_cached = seq_len\n \n if seq_len < self.original_max_seq_len and self.max_seq_len_cached > self.original_max_seq_len: # ...
2025-01-13T09:11:17
golang/go
382851c1fd135a99efbe128a3be0ce466d42506f
9f5298ca6e7fc9c46c0a82bd7be39450ec48dcb5
cmd/compile: fix failure to communicate between ABIinfo producer&consumer ABI info producer and consumer had different ideas for register order for parameters. Includes a test, includes improvements to debugging output. Updates #44816. Change-Id: I4812976f7a6c08d6fc02aac1ec0544b1f141cca6 Reviewed-on: https://go-rev...
[ { "path": "src/cmd/compile/internal/abi/abiutils.go", "patch": "@@ -477,29 +477,35 @@ func (c *RegAmounts) regString(r RegIndex) string {\n \treturn fmt.Sprintf(\"<?>%d\", r)\n }\n \n-// toString method renders an ABIParamAssignment in human-readable\n+// ToString method renders an ABIParamAssignment in hum...
2021-03-07T19:00:10
ollama/ollama
9fa80a1660bc8ba28a77e6e91cc14d1c2b1cb652
dde09129d143c99e3f78859d3829d215501aa584
app/ui: fix lint errors for unused vars, prefer-const, and empty catch (#15282)
[ { "path": "app/ui/app/src/components/Message.tsx", "patch": "@@ -536,7 +536,7 @@ function ToolCallDisplay({\n let args: Record<string, unknown> | null = null;\n try {\n args = JSON.parse(toolCall.function.arguments) as Record<string, unknown>;\n- } catch (e) {\n+ } catch {\n args =...
2026-04-07T23:28:36
vercel/next.js
a012f64a3c0b8c23661ff85734cbdbeb5c06f409
352e2ad3c52fad93de1e47cf8a6f3ffb305d46c4
error overlay (#39) transferred from https://github.com/vercel/turbo-tooling/pull/396 Co-authored-by: Tobias Koppers <tobias.koppers@googlemail.com>
[ { "path": "crates/next-core/js/build.mjs", "patch": "@@ -0,0 +1,162 @@\n+/**\n+ * Build npm package to be able to embed them in the binary\n+ */\n+\n+import { mkdir, writeFile, rm, readFile } from \"node:fs/promises\";\n+import { existsSync } from \"node:fs\";\n+import { createRequire } from \"node:module\"...
2022-10-20T16:04:19
rust-lang/rust
547a31016d9725df7466f6064fdcf0e4fb031957
4a8026ce639f2e5e09d20320ca569fedcafcef6a
Rename some `name` variables as `ident`. It bugs me when variables of type `Ident` are called `name`. It leads to silly things like `name.name`. `Ident` variables should be called `ident`, and `name` should be used for variables of type `Symbol`. This commit improves things by by doing `s/name/ident/` on a bunch of `...
[ { "path": "src/main_shim.rs", "patch": "@@ -104,7 +104,7 @@ pub(crate) fn maybe_create_entry_wrapper(\n let termination_trait = tcx.require_lang_item(LangItem::Termination, None);\n let report = tcx\n .associated_items(termination_trait)\n- ...
2025-04-08T02:23:07
huggingface/transformers
1e3c6c1f7d7fbb4862ff036a9bdf96e6ae2e2511
04eae987f3f856724fddba9d6f9afbcacfdfed27
Skip `MobileNetV1ModelTest::test_batching_equivalence` for now (#35614) fix Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
[ { "path": "tests/models/mobilenet_v1/test_modeling_mobilenet_v1.py", "patch": "@@ -214,6 +214,8 @@ def test_model_from_pretrained(self):\n model = MobileNetV1Model.from_pretrained(model_name)\n self.assertIsNotNone(model)\n \n+ # TODO: ydshieh\n+ @unittest.skip(\"skip for now as #35564...
2025-01-10T17:32:36
golang/go
9f5298ca6e7fc9c46c0a82bd7be39450ec48dcb5
98dfdc82c85c238a5ab6131d6f1e86e0da259851
cmd/compile: fix confusion in generating SelectN index Old: return the ABI register index of the result (wrong!) New: return the index w/in sequence of result registers (right!) Fixed bug: genCaller0/genCaller0.go:43:9: internal compiler error: 'Caller0': panic during schedule while compiling Caller0: runtime erro...
[ { "path": "src/cmd/compile/internal/ssa/expand_calls.go", "patch": "@@ -424,7 +424,11 @@ func (x *expandState) rewriteSelect(leaf *Value, selector *Value, offset int64,\n \t\t\t\t}\n \t\t\t\toutParam := aux.abiInfo.OutParam(int(which))\n \t\t\t\tif len(outParam.Registers) > 0 {\n-\t\t\t\t\treg := int64(outP...
2021-03-07T01:59:40
vercel/next.js
6d29713023467f6aa9d0aa2e1d7a67fc4a26c4fc
12408a3f2df217655f1ca1f5a63e437776208a47
perf: refactor path logic in router + add LRU cache (#41365) I'm investigating the runtime perf of the node server and this was one of the hot spot I stumbled upon. This diff: - refactors getPagePath to not throw when it doesn't find a path: this function is used to check if a path exists, we don't want that kind of ...
[ { "path": "packages/next/server/next-server.ts", "patch": "@@ -74,7 +74,7 @@ import BaseServer, {\n NoFallbackError,\n RequestContext,\n } from './base-server'\n-import { getPagePath, requireFontManifest } from './require'\n+import { getMaybePagePath, getPagePath, requireFontManifest } from './require'\...
2022-10-20T15:42:50
ollama/ollama
30fdd229a434cfae409cc07456684315bf95a561
e823bff87379a8d6dd1927c22830e9ffda6af130
create: Clean up experimental paths, fix create from existing safetensor model (#14679) * create: Clean up experimental paths This cleans up the experimental features, and adds both unit and integration test coverage to verify no regressions. * create: preserve config and layer names when creating from safetensors...
[ { "path": ".gitignore", "patch": "@@ -15,3 +15,4 @@ __debug_bin*\n llama/build\n llama/vendor\n /ollama\n+integration/testdata/models/", "additions": 1, "deletions": 0, "language": "Unknown" }, { "path": "cmd/cmd.go", "patch": "@@ -54,7 +54,6 @@ import (\n \t\"github.com/ollama/ollam...
2026-04-07T15:12:57
huggingface/transformers
04eae987f3f856724fddba9d6f9afbcacfdfed27
b02828e4af74373c97c03a27e2921942b7eb8557
Fix flaky `test_beam_search_low_memory` (#35611) * fix * fix * fix * fix * fix --------- Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
[ { "path": "tests/generation/test_utils.py", "patch": "@@ -204,6 +204,8 @@ def _get_logits_processor_kwargs(self, do_sample=False, config=None):\n \"vision_start_token_id\",\n ]:\n token_index = getattr(config, key, None)\n+ if token_index is None an...
2025-01-10T16:31:03
rust-lang/rust
1b3fc585cb48c71dbaf1677af6d6698977b436cd
d4f880f8ce832cd7560bb2f1ebc34f967055ffd7
Rename some `name` variables as `ident`. It bugs me when variables of type `Ident` are called `name`. It leads to silly things like `name.name`. `Ident` variables should be called `ident`, and `name` should be used for variables of type `Symbol`. This commit improves things by by doing `s/name/ident/` on a bunch of `...
[ { "path": "compiler/rustc_ast/src/expand/mod.rs", "patch": "@@ -13,12 +13,12 @@ pub mod typetree;\n #[derive(Debug, Clone, Encodable, Decodable, HashStable_Generic)]\n pub struct StrippedCfgItem<ModId = DefId> {\n pub parent_module: ModId,\n- pub name: Ident,\n+ pub ident: Ident,\n pub cfg: Me...
2025-04-08T02:23:07
golang/go
98dfdc82c85c238a5ab6131d6f1e86e0da259851
5eb99120844c0494d655678262e1fb41949a2b99
cmd/compile: fix broken type+offset calc for register args Includes more enhancements to debugging output. Updates #44816. Change-Id: I5b21815cf37ed21e7dec6c06f538090f32260203 Reviewed-on: https://go-review.googlesource.com/c/go/+/299409 Trust: David Chase <drchase@google.com> Reviewed-by: Cherry Zhang <cherryyz@goo...
[ { "path": "src/cmd/compile/internal/abi/abiutils.go", "patch": "@@ -130,10 +130,15 @@ func (pa *ABIParamAssignment) RegisterTypesAndOffsets() ([]*types.Type, []int64)\n \t}\n \ttyps := make([]*types.Type, 0, l)\n \toffs := make([]int64, 0, l)\n-\treturn appendParamTypes(typs, pa.Type), appendParamOffsets(of...
2021-03-06T02:09:40
vercel/next.js
d7348c4ef42c649f08417d09234e689bd9f5fcff
a5d53155cddb41fec1901955bb660fa6c1d4b4cf
Full remaining path in selected layout segment (#41562) Make `useSelectedLayoutSegment` include the remaining segments from the current level to the leaf node. ## Bug - [ ] Related issues linked using `fixes #number` - [ ] Integration tests added - [ ] Errors have a helpful link attached, see `contributing.md` ## F...
[ { "path": "packages/next/client/components/navigation.ts", "patch": "@@ -1,5 +1,6 @@\n // useLayoutSegments() // Only the segments for the current place. ['children', 'dashboard', 'children', 'integrations'] -> /dashboard/integrations (/dashboard/layout.js would get ['children', 'dashboard', 'children', 'in...
2022-10-20T11:45:16
ollama/ollama
8c8f8f3450d39735355fc6cd7f2e436c8aa42ab1
82f01395872705e531845b59e2ea17f7b1fd4890
model/parsers: add gemma4 tool call repair (#15374) The existing strict gemma4 tool parser is still the primary path, but if this fails, we try to repair by fixing some of the most commonly seen mistakes these models seem to make in practice. We repair by building up a set of candidates, and use the first candidate t...
[ { "path": "model/parsers/gemma4.go", "patch": "@@ -7,6 +7,7 @@ import (\n \t\"regexp\"\n \t\"strings\"\n \t\"unicode\"\n+\t\"unicode/utf8\"\n \n \t\"github.com/ollama/ollama/api\"\n )\n@@ -25,6 +26,7 @@ const (\n \tgemma4ThinkingCloseTag = \"<channel|>\"\n \tgemma4ToolCallOpenTag = \"<|tool_call>\"\n \tgem...
2026-04-07T01:47:17
rust-lang/rust
77116881a857ca49ff11c7c3b35c86305864ab45
934880f586f6ac1f952c7090e2a943fcd7775e7b
update miniz_oxide to 0.8.8 0.8.7 can trigger a panic when debug assertions are enabled when used via flate2 in some cases
[ { "path": "Cargo.lock", "patch": "@@ -1221,7 +1221,7 @@ source = \"registry+https://github.com/rust-lang/crates.io-index\"\n checksum = \"7ced92e76e966ca2fd84c8f7aa01a4aea65b0eb6648d72f7c8f3e2764a67fece\"\n dependencies = [\n \"crc32fast\",\n- \"miniz_oxide 0.8.7\",\n+ \"miniz_oxide 0.8.8\",\n ]\n \n [[pac...
2025-04-09T21:45:46
huggingface/transformers
0aaf124fb934c314fe28fa9d1fa935c39d69f911
1211e616a44fbfa864b6e196219b5b54dfd07aeb
Added error when sequence length is bigger than max_position_embeddings (#32156) * Added error when sequence length is bigger than max_position_embeddings * Fixed formatting * Fixed bug * Changed copies to match * Fixed bug * Applied suggestions * Removed redundant code * Fixed bugs * Bug fix ...
[ { "path": "src/transformers/models/blip/modeling_blip.py", "patch": "@@ -309,6 +309,13 @@ def forward(\n inputs_embeds: Optional[torch.FloatTensor] = None,\n ) -> torch.Tensor:\n seq_length = input_ids.shape[-1] if input_ids is not None else inputs_embeds.shape[-2]\n+ max_position...
2025-01-10T15:23:54
nodejs/node
f413f56c3606fa6f01f0a7341fb4136965890fb7
5c425788f1087f4f41fb6174692433d1bf4ce891
util: fix inspect performance bug In case an object contained a circular reference `Object.keys` was called even though it was not necessary at all. This caused a significant overhead for objects that contained a lot of such entries. PR-URL: https://github.com/nodejs/node/pull/20007 Reviewed-By: Matteo Collina <matte...
[ { "path": "lib/util.js", "patch": "@@ -456,6 +456,11 @@ function formatValue(ctx, value, recurseTimes, ln) {\n }\n }\n \n+ // Using an array here is actually better for the average case than using\n+ // a Set. `seen` will only check for the depth and will never grow too large.\n+ if (ctx.seen.index...
2018-04-13T12:21:45
golang/go
5eb99120844c0494d655678262e1fb41949a2b99
48895d021bf631f15d68ecc10cab89ebd9cb28f6
cmd/compile: fix OpArg decomposer for registers in expandCalls Includes test taken from https://github.com/golang/go/issues/44816#issuecomment-791618179 and improved debugging output. Updates #44816 Change-Id: I94aeb9c5255f175fe80727be29d218bad54bf7ea Reviewed-on: https://go-review.googlesource.com/c/go/+/299389 Tru...
[ { "path": "src/cmd/compile/internal/ssa/expand_calls.go", "patch": "@@ -78,7 +78,8 @@ func (rc *registerCursor) String() string {\n \t\t\tregs = regs + x.LongString()\n \t\t}\n \t}\n-\treturn fmt.Sprintf(\"RCSR{storeDest=%v, regsLen=%d, nextSlice=%d, regValues=[%s], config=%v\", dest, rc.regsLen, rc.nextSli...
2021-03-06T00:56:13
vercel/next.js
a5d53155cddb41fec1901955bb660fa6c1d4b4cf
765791a6d1fe56bf4247ea676f4ef8a0f803c034
Check root layout change on client (#41475) Moves the logic that checks if there's a new root layout to the client. Adds test for static and dynamic catchall. Related: #41457 ## Bug - [ ] Related issues linked using `fixes #number` - [ ] Integration tests added - [ ] Errors have a helpful link attached, se...
[ { "path": "packages/next/client/components/reducer.ts", "patch": "@@ -406,7 +406,7 @@ function applyRouterStatePatchToTree(\n flightRouterState: FlightRouterState,\n treePatch: FlightRouterState\n ): FlightRouterState | null {\n- const [segment, parallelRoutes /* , url */] = flightRouterState\n+ const...
2022-10-20T01:38:00
ollama/ollama
3cd2b03a5e81f39b132ccd95b8b250f158f97981
c8e0878814b4d19200d65571d3d2d35b4b48fd3e
ggml: fix ROCm build for cublasGemmBatchedEx reserve wrapper Add missing cublasGemmAlgo_t to hipblasGemmAlgo_t type mapping and cast away const qualifiers that hipblasGemmBatchedEx doesn't accept.
[ { "path": "llama/patches/0020-ggml-No-alloc-mode.patch", "patch": "@@ -229,7 +229,7 @@ diff --git a/ggml/src/ggml-cuda/common.cuh b/ggml/src/ggml-cuda/common.cuh\n index 9fcb2f9fd..e800ee8f6 100644\n --- a/ggml/src/ggml-cuda/common.cuh\n +++ b/ggml/src/ggml-cuda/common.cuh\n-@@ -37,6 +37,62 @@\n+@@ -37,6 +3...
2026-04-03T20:26:50
huggingface/transformers
1211e616a44fbfa864b6e196219b5b54dfd07aeb
bbc00046b9b746d5ea69a27a58e652fdf617c91c
Use inherit tempdir makers for tests + fix failing DS tests (#35600) * Use existing APIs to make tempdir folders * Fixup deepspeed too * output_dir -> tmp_dir
[ { "path": "tests/deepspeed/test_deepspeed.py", "patch": "@@ -482,6 +482,7 @@ def test_hf_ds_config_mismatch(self):\n max_grad_norm=max_grad_norm,\n adam_beta1=adam_beta1,\n adam_beta2=adam_beta2,\n+ output_dir=self.get_auto_remove_tmp_dir(),\n ...
2025-01-10T15:01:58
nodejs/node
5c425788f1087f4f41fb6174692433d1bf4ce891
f64bebf2059d35299da58cf9c5ca22d68035d617
errors: alter ERR_INVALID_PROTOCOL Changes the base instance for ERR_INVALID_PROTOCOL from Error to TypeError as a more accurate representation of the error. PR-URL: https://github.com/nodejs/node/pull/19983 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Michaël Zasso <targos@protonmail.com> Reviewed-By:...
[ { "path": "lib/internal/errors.js", "patch": "@@ -896,9 +896,9 @@ E('ERR_INVALID_OPT_VALUE_ENCODING',\n 'The value \"%s\" is invalid for option \"encoding\"', TypeError);\n E('ERR_INVALID_PERFORMANCE_MARK',\n 'The \"%s\" performance mark has not been set', Error);\n-\n-// This should probably be a `Type...
2018-04-12T18:37:11
golang/go
6ee1506769b24d99e27f9a6a9c99e9b7143112bf
d4825819fe58e8531c7fcdf4ce27cec63824db25
[dev.fuzz] internal/fuzz: fix deadlock with multiple workers CoordinateFuzzing now continues to run after discovering a crasher. It waits until all workers have terminated before returning. This fixes a deadlock that occurred when multiple workers discovered crashers concurrently. CoordinateFuzzing would receive one ...
[ { "path": "src/cmd/go/testdata/script/test_fuzz.txt", "patch": "@@ -9,12 +9,12 @@ stdout FAIL\n \n # Test that fuzzing a fuzz target that returns without failing or calling\n # f.Fuzz fails and causes a non-zero exit status.\n-! go test -fuzz=Fuzz -fuzztime=5s -parallel=1 noop_fuzz_test.go\n+! go test -fuzz...
2021-02-17T22:31:22
huggingface/transformers
bbc00046b9b746d5ea69a27a58e652fdf617c91c
f63829c87bd89a4a0cea45d81c1cd870996b30c4
Fix flaky `test_custom_4d_attention_mask` (#35606) * fix * fix --------- Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
[ { "path": "src/transformers/testing_utils.py", "patch": "@@ -1431,14 +1431,20 @@ def set_model_tester_for_less_flaky_test(test_case):\n and target_num_hidden_layers is not None\n ):\n test_case.model_tester.vision_config = copy.deepcopy(test_case.model_tester.vision_config)\n- tes...
2025-01-10T14:40:04
ollama/ollama
bb0c58e13416574813361d27ddf4699bf24d3090
036ed1b9b50558907789c66b407a6cd05d31cc20
ggml: skip cublasGemmBatchedEx during graph reservation cublasGemmBatchedEx fails during graph capture when pool allocations return fake pointers. This is triggered when NUM_PARALLEL is greater than 1 for models like gemma4 that use batched matmuls. Skip it during reservation since the memory tracking is already handl...
[ { "path": "llama/patches/0020-ggml-No-alloc-mode.patch", "patch": "@@ -11,9 +11,9 @@ must be recreated with no-alloc set to false before loading data.\n ggml/include/ggml-backend.h | 1 +\n ggml/src/ggml-backend-impl.h | 16 +++\n ggml/src/ggml-backend.cpp | 75 ++++++++++-\n- ggml/src/ggml...
2026-04-03T18:26:03
vercel/next.js
765791a6d1fe56bf4247ea676f4ef8a0f803c034
b0db95ffc612f991b19f9746adf6653785a33bec
Support overriding request headers in middlewares (#41380) This PR adds a feature in middleware to add, modify, or delete request headers. This feature is quite useful to pass data from middleware to Serverless/Edge API routes. ### Questions for Reviewers - Should we deny modifying standard request headers like `T...
[ { "path": "packages/next/server/next-server.ts", "patch": "@@ -1882,6 +1882,37 @@ export default class NextNodeServer extends BaseServer {\n result.response.headers.set('x-middleware-rewrite', rel)\n }\n \n+ if (result.response.headers.has('x-middleware-override-headers'...
2022-10-20T01:36:05
rust-lang/rust
ccdfd310be12f0c46be8b3266b3ff4e2ce5b3806
27836e1e5794dfa8db8565adaf7aed05d8bfff92
Mark GAT WC as GoalSource::AliasWellFormed so that we recurse into them in error reporting
[ { "path": "compiler/rustc_next_trait_solver/src/solve/normalizes_to/mod.rs", "patch": "@@ -134,7 +134,7 @@ where\n // Add GAT where clauses from the trait's definition\n // FIXME: We don't need these, since these are the type's own WF obligations.\n ...
2025-03-26T18:59:44
nodejs/node
f7fbbeedc609f56c898230971b44d3dde0934dc9
cf48d1db660888cd3bfd14953a27527d56e90e6a
http: relax requirements on upgrade listener The http spec does not say anything about Upgrade headers making protocol switch mandatory but Node.js implements them as if they are. Relax the requirements to only destroy the socket if no upgrade listener exists on the client when status code is 101. PR-URL: https://git...
[ { "path": "doc/api/http.md", "patch": "@@ -471,8 +471,9 @@ added: v0.1.94\n * `head` {Buffer}\n \n Emitted each time a server responds to a request with an upgrade. If this\n-event is not being listened for, clients receiving an upgrade header will have\n-their connections closed.\n+event is not being liste...
2018-04-12T17:57:19
golang/go
d4825819fe58e8531c7fcdf4ce27cec63824db25
ac58614190c9dd1ca0a1a47bf7a29a4a7c501e72
[dev.fuzz] internal/fuzz: fix two bugs affecting windows * Appending to the worker environment slice should reallocate it. On Windows, we pass handles through the environment, and concurrent workers were writing to the same memory, resulting in "The handle is invalid" errors. * Instead of passing a handle to the...
[ { "path": "src/internal/fuzz/fuzz.go", "patch": "@@ -96,7 +96,7 @@ func CoordinateFuzzing(ctx context.Context, parallel int, seed []CorpusEntry, ty\n \t\t\tdir: dir,\n \t\t\tbinPath: binPath,\n \t\t\targs: args,\n-\t\t\tenv: env,\n+\t\t\tenv: env[:len(env):len(env)], // co...
2021-02-26T17:57:44
ollama/ollama
036ed1b9b50558907789c66b407a6cd05d31cc20
3536ef58f613b9f448a040b09417aa057e52bc2d
model/parsers: fix gemma4 arg parsing when quoted strings contain " (#15254) * model/parsers: fix gemma4 arg parsing when quoted strings contain " Fixes: #15241 * add more tests, be careful about what we escape We want Windows-style paths to not get misinterpreted * fix backslash-quote case, it really should be a ...
[ { "path": "model/parsers/gemma4.go", "patch": "@@ -345,27 +345,85 @@ func parseGemma4ToolCall(content string) (api.ToolCall, error) {\n \n // gemma4ArgsToJSON converts Gemma 4's custom argument format to valid JSON.\n func gemma4ArgsToJSON(s string) string {\n-\ts = strings.ReplaceAll(s, `<|\"|>`, `\"`)\n+\...
2026-04-03T05:52:51
rust-lang/rust
199ee4084349361ae44c6e4b0cbc9a5e72913b20
48f89e7659678f91a68c0c2d868180a0036ab32d
Move errors
[ { "path": "compiler/rustc_trait_selection/src/error_reporting/traits/mod.rs", "patch": "@@ -2,6 +2,7 @@ pub mod ambiguity;\n pub mod call_kind;\n mod fulfillment_errors;\n pub mod on_unimplemented;\n+pub mod on_unimplemented_format;\n mod overflow;\n pub mod suggestions;\n ", "additions": 1, "deleti...
2025-03-25T13:48:12
nodejs/node
345e3b28c6aee09b86311f5bcc23fbe818f1ea4c
978e1524bba34fb92e6710f589329cd03cbec87b
errors: alter ERR_INVALID_DOMAIN_NAME Changes the base instance for ERR_INVALID_DOMAIN_NAME from Error to TypeError as a more accurate representation of the error. PR-URL: https://github.com/nodejs/node/pull/19961 Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Revi...
[ { "path": "lib/internal/errors.js", "patch": "@@ -878,9 +878,7 @@ E('ERR_INVALID_CHAR', invalidChar, TypeError);\n // This should probably be a `TypeError`.\n E('ERR_INVALID_CURSOR_POS',\n 'Cannot set cursor row without setting its column', Error);\n-\n-// This should probably be a `TypeError`.\n-E('ERR_I...
2018-04-12T00:34:59
golang/go
ac58614190c9dd1ca0a1a47bf7a29a4a7c501e72
f251d1fbb69156627379df28a150343c08a29474
[dev.fuzz] testing: use exit status 70 for worker errors (not crashes) If a worker process encounters an error communicating with the coordinator, or if the setup code reports an error with F.Fail before calling F.Fuzz, exit with status 70. The coordinator will report these errors and 'go test' will exit non-zero, but...
[ { "path": "src/cmd/go/testdata/script/test_fuzz_mutate_fail.txt", "patch": "@@ -0,0 +1,103 @@\n+# TODO(jayconrod): support shared memory on more platforms.\n+[!darwin] [!linux] [!windows] skip\n+\n+# Check that if a worker does not call F.Fuzz or calls F.Fail first,\n+# 'go test' exits non-zero and no crash...
2021-02-26T17:50:48
vercel/next.js
b0db95ffc612f991b19f9746adf6653785a33bec
cec9d02dee2ab2c6d100ad3cbcafa4c8d6ab4805
Update error message for invalid react-dom/server imports (#41582) Related discussion: https://vercel.slack.com/archives/C043ANYDB24/p1666208186363099?thread_ts=1664676119.265829&cid=C043ANYDB24. Tl;dr: RSC already renders and serializes things dynamically, there is no need to manually do that in RSC and embed th...
[ { "path": "packages/next/build/webpack/plugins/wellknown-errors-plugin/parseRSC.ts", "patch": "@@ -24,10 +24,16 @@ export function formatRSCErrorMessage(\n formattedVerboseMessage =\n '\\n\\nMaybe one of these should be marked as a client entry with \"use client\":\\n'\n } else if (NEXT_RS...
2022-10-20T00:56:14
huggingface/transformers
52e1f87c7d2fdaf147aa625c169ad440672cb3cb
ccc0381d36aa99000311b708382165558784c27d
[WIP] Emu3: add model (#33770) * model can convert to HF and be loaded back * nit * works in single batch generation but hallucinates * use the image tokens * add image generation * now it works * add tests * update * add modulare but it doesn't work for porting docstring :( * skip some test...
[ { "path": "docs/source/en/_toctree.yml", "patch": "@@ -860,6 +860,8 @@\n title: DePlot\n - local: model_doc/donut\n title: Donut\n+ - local: model_doc/emu3\n+ title: Emu3\n - local: model_doc/flava\n title: FLAVA\n - local: model_doc/git", "additions...
2025-01-10T11:23:00
nodejs/node
978e1524bba34fb92e6710f589329cd03cbec87b
2b6bb9f588aa791a385c8a5075ab5e207f9a739d
timers: fix clearInterval to work with timers from setTimeout According to HTML Living Standard, "either method [clearInterval or clearTimeout] can be used to clear timers created by setTimeout() or setInterval().". The current implementation of clearTimeout is already able to destroy a timer created by setInterval, ...
[ { "path": "doc/api/timers.md", "patch": "@@ -56,8 +56,8 @@ Returns a reference to the `Immediate`.\n ## Class: Timeout\n \n This object is created internally and is returned from [`setTimeout()`][] and\n-[`setInterval()`][]. It can be passed to [`clearTimeout()`][] or\n-[`clearInterval()`][] (respectively) ...
2018-04-11T15:49:07
golang/go
f251d1fbb69156627379df28a150343c08a29474
b89483497a7349bb8dba9110e765f5ff1189f69d
[dev.fuzz] testing, internal/fuzz: multiple small fixes * Run gofmt with go1.17 build constraint changes. * Tighten regular expressions used in tests. "ok" got some false positives with verbose output, so make sure it appears at the start of a line. * Return err in deps.RunFuzzWorker instead of nil. * Call common....
[ { "path": "src/cmd/go/internal/test/genflags.go", "patch": "@@ -2,6 +2,7 @@\n // Use of this source code is governed by a BSD-style\n // license that can be found in the LICENSE file.\n \n+//go:build ignore\n // +build ignore\n \n package main", "additions": 1, "deletions": 0, "language": "Go" ...
2021-02-26T15:54:11
vercel/next.js
4adab6a61e2e624f26e6685fc67dc2511e97e135
bad909e4d16d7292878bb0209538adf6ddcb3711
Improve server bundling strategy (#41584) This PR adds a list of popular packages that should always opt-out bundling in the server layer. ## Bug - [ ] Related issues linked using `fixes #number` - [ ] Integration tests added - [ ] Errors have a helpful link attached, see `contributing.md` ## Feature - ...
[ { "path": "packages/next/build/webpack-config.ts", "patch": "@@ -13,6 +13,7 @@ import {\n WEBPACK_LAYERS,\n RSC_MOD_REF_PROXY_ALIAS,\n } from '../lib/constants'\n+import { EXTERNAL_PACKAGES } from '../lib/server-external-packages'\n import { fileExists } from '../lib/file-exists'\n import { CustomRoutes...
2022-10-20T00:45:37
ollama/ollama
de9673ac3fb1c57fbf6e5e194f1f3dc5a8b48668
96b202d34b82d1755887bf4204e1f2e053720d4f
tokenizer: add byte fallback for SentencePiece BPE encoding (#15232) * tokenizer: add byte fallback for SentencePiece BPE encoding When BPE merging produces tokens not in the vocabulary, fall back to encoding each UTF-8 byte as <0xHH> byte tokens instead of silently dropping the character. Also teach Decode to conver...
[ { "path": "model/models/gemma4/tokenizer_reference_test.go", "patch": "@@ -0,0 +1,341 @@\n+package gemma4\n+\n+// TestGemma4TokenizerMatchesReference verifies our BPE tokenizer matches\n+// the Rust tokenizers library (the reference implementation) for Gemma 4.\n+//\n+// The test loads vocabulary from any l...
2026-04-02T20:04:45
huggingface/transformers
ccc0381d36aa99000311b708382165558784c27d
a9bd1e6284429bab46bdf73620f90a0a22bcf75e
Fix flex_attention in training mode (#35605) * fix flex * add test * style
[ { "path": "src/transformers/integrations/flex_attention.py", "patch": "@@ -27,7 +27,7 @@ def causal_mod(score, b, h, q_idx, kv_idx):\n if softcap is not None:\n score = softcap * torch.tanh(score / softcap)\n if causal_mask is not None:\n- score += causal_mask[b][0][q_...
2025-01-10T10:49:12
nodejs/node
2b6bb9f588aa791a385c8a5075ab5e207f9a739d
de66bd34b9bbaab16492ac4f4940006a02b14aee
deps: c-ares float, win ipv6 bad fec0 prefix Was 72c5458aee: PR-URL: https://github.com/nodejs/node/pull/5090 Reviewed-By: Fedor Indutny <fedor@indutny.com> Reimplemented for c-ares 1.13.0 PR-URL: https://github.com/nodejs/node/pull/15378 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Ruben Bridgew...
[ { "path": "deps/cares/src/ares_init.c", "patch": "@@ -1272,6 +1272,11 @@ static int get_DNS_AdaptersAddresses(char **outptr)\n }\n else if (namesrvr.sa->sa_family == AF_INET6)\n {\n+ /* Windows apparently always reports some IPv6 DNS servers that\n+ * prefixed with fec0:0:0:...
2013-05-15T19:25:11
golang/go
e8e425cb231863c1ed1bcf00dd5a517ceeaf6b9c
b60a3a8cfbc6096babe72ebcee0733bba496cf9a
runtime: add pollDesc partial edges gscan is taken during stack growth, which may occur while pollDesc is held. mallocgc may also be called while pollDesc is held. mallocgc may take mheap or mheapSpecial. The former exists, but is out of order; the latter is missing. Fixes #44881 Change-Id: Ie25935d9d433e813c11a528...
[ { "path": "src/runtime/lockrank.go", "patch": "@@ -224,14 +224,14 @@ var lockPartialOrder [][]lockRank = [][]lockRank{\n \tlockRankRwmutexR: {lockRankSysmon, lockRankRwmutexW},\n \n \tlockRankSpanSetSpine: {lockRankSysmon, lockRankScavenge, lockRankForcegc, lockRankAssistQueue, lockRankCpuprof, lockRankSwee...
2021-03-09T15:41:16
vercel/next.js
bad909e4d16d7292878bb0209538adf6ddcb3711
d261e7cd253f57efdcc35bfc7b9fd1a27f20588f
Update page config APIs (#41580) This PR updates app configurations from `export const config = { ... }` to be directly exported. ## Bug - [ ] Related issues linked using `fixes #number` - [ ] Integration tests added - [ ] Errors have a helpful link attached, see `contributing.md` ## Feature - [ ] Imple...
[ { "path": "packages/next/build/analysis/get-page-static-info.ts", "patch": "@@ -48,22 +48,39 @@ export function getRSCModuleType(source: string): RSCModuleType {\n * requires a runtime to be specified. Those are:\n * - Modules with `export function getStaticProps | getServerSideProps`\n * - Modules w...
2022-10-20T00:39:25
huggingface/transformers
e0646f3dce229a2552c56746eac618baf8210efa
5f087d1335e327187ab4edec746d3d5339a68a79
Chat template: return vectorized output in processors (#34275) * update chat template * style * fix tests * Update src/transformers/image_utils.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * typehints + docs * fix tests * remove unnecessary warnings * forgot code style :( * allow ...
[ { "path": "benchmark.py", "patch": "@@ -0,0 +1,132 @@\n+import os\n+import time\n+\n+import cv2\n+import av\n+import numpy as np \n+from numba import jit, cuda\n+from decord import VideoReader, cpu, gpu\n+\n+import torch\n+from torchvision import io\n+\n+\n+video_dir = \"/raid/raushan/temp_dir/\"\n+NUM_FRAM...
2025-01-10T10:05:29
ollama/ollama
96b202d34b82d1755887bf4204e1f2e053720d4f
79865e6c5a2f8aa7bc8135eacffdcbe2fea953d9
Add support for gemma4 (#15214) * bench: add prompt calibration, context size flag, and NumCtx reporting Add --num-ctx flag to set context size, and report NumCtx in model info header. Calibrate tokens-per-word ratio during warmup using actual tokenization metrics from the model, replacing the fixed 1.3 heuristic. Th...
[ { "path": "api/types.go", "patch": "@@ -436,6 +436,7 @@ type ToolProperty struct {\n \tDescription string `json:\"description,omitempty\"`\n \tEnum []any `json:\"enum,omitempty\"`\n \tProperties *ToolPropertiesMap `json:\"properties,omitempty\"`\n+\tRequired []string ...
2026-04-02T18:33:33
nodejs/node
b0b7afcfbe0f38715af062ebcb9e1c6bbc68ba68
4a0dff785e766404874c4f171efa85c9508aac5e
deps: upgrade to c-ares v1.14.0 * Fix patch for CVE-2017-1000381 to not be overly aggressive. * win32: Preserve DNS server order returned by Windows when sorting and exclude DNS servers in legacy subnets. * win32: Support most recent Visual Studio 2017. * gethostbyaddr should fail with ECANCELLED not ENOTFOUND ...
[ { "path": "deps/cares/cares.gyp", "patch": "@@ -40,6 +40,7 @@\n 'include/ares_rules.h',\n 'include/ares_version.h',\n 'include/nameser.h',\n+ 'src/ares_android.c',\n 'src/ares_cancel.c',\n 'src/ares__close_sockets.c',\n 'src/ares_create_query.c',", ...
2018-04-11T11:34:43
golang/go
b60a3a8cfbc6096babe72ebcee0733bba496cf9a
a70eb2c9f2e3d52116b90a21ce10718356197ee5
cmd/compile: add debugging mode for import/export Just add a simple magic number with each op, to detect when the reader gets desynchronized from the writer. Change-Id: Iac7dab7f465b0021b1d7ae31c8f8a353ac3663a2 Reviewed-on: https://go-review.googlesource.com/c/go/+/299769 Trust: Keith Randall <khr@golang.org> Run-Try...
[ { "path": "src/cmd/compile/internal/typecheck/iexport.go", "patch": "@@ -246,6 +246,11 @@ const (\n \tinterfaceType\n )\n \n+const (\n+\tdebug = false\n+\tmagic = 0x6742937dc293105\n+)\n+\n func WriteExports(out *bufio.Writer) {\n \tp := iexporter{\n \t\tallPkgs: map[*types.Pkg]bool{},\n@@ -1584,6 +1589...
2021-03-05T17:49:28
vercel/next.js
d261e7cd253f57efdcc35bfc7b9fd1a27f20588f
2551978fa40b68882daa349019f13f8f783571f5
Add back() and forward() to new router (#41575) Adds `router.back()` and `router.forward()`. <!-- Thanks for opening a PR! Your contribution is much appreciated. To make sure your PR is handled as smoothly as possible we request that you follow the checklist sections below. Choose the right checklist for the chan...
[ { "path": "packages/next/client/components/app-router.tsx", "patch": "@@ -203,6 +203,8 @@ function Router({\n }\n \n const routerInstance: AppRouterInstance = {\n+ back: () => window.history.back(),\n+ forward: () => window.history.forward(),\n // TODO-APP: implement prefetching of f...
2022-10-20T00:35:43
huggingface/transformers
5f087d1335e327187ab4edec746d3d5339a68a79
6f127d3f81c8cf434aa809a4d8ec76b3a6372060
Add Moonshine (#34784) * config draft * full encoder forward * full decoder forward * fix sdpa and FA2 * fix sdpa and FA2 * moonshine model * moonshine model forward * fix attention with past_key_values * add MoonshineForConditionalGeneration * fix cache handling and causality for cross att...
[ { "path": "docs/source/en/_toctree.yml", "patch": "@@ -505,7 +505,9 @@\n - local: model_doc/mobilebert\n title: MobileBERT\n - local: model_doc/modernbert\n- title: ModernBERT\n+ title: ModernBert\n+ - local: model_doc/moonshine\n+ title: moonshine\n - loc...
2025-01-10T10:00:54
ollama/ollama
9e7cb9697edf3782a0f763dab1a36985ae0ff6a5
3824e380a82d3368855cc1c5b9d128cc2849c4f6
mlx: fix vision capability + min version (#15106)
[ { "path": "cmd/cmd_test.go", "patch": "@@ -301,7 +301,7 @@ Weigh anchor!\n \t\t\t\tParameterSize: \"7B\",\n \t\t\t\tQuantizationLevel: \"FP16\",\n \t\t\t},\n-\t\t\tRequires: \"0.14.0\",\n+\t\t\tRequires: \"0.19.0\",\n \t\t}, false, &b); err != nil {\n \t\t\tt.Fatal(err)\n \t\t}\n@@ -310,10 +310,17 @@ We...
2026-03-28T00:09:28
nodejs/node
1d6cdfce807ba81f372cd3ab6d6ba81f71abeb6b
f7496896c4ea088cbd5e4ed29eeecca655e1ac0b
doc: improve assert documentation This adds a example to `assert.throws` to document checking against error instances. It also makes it clearer what some arguments can receive as input. PR-URL: https://github.com/nodejs/node/pull/19885 Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Michaël Zasso ...
[ { "path": "doc/api/assert.md", "patch": "@@ -395,6 +395,9 @@ is little benefit by catching a rejection and then rejecting it again. Instead,\n consider adding a comment next to the specific code path that should not reject\n and keep error messages as expressive as possible.\n \n+If specified, `error` can b...
2018-04-08T23:31:59
golang/go
762ef81a5670da99a4b060fd3afe4134c9e82017
b70a2bc9c612de35b765712bd689865f6a1716b6
cmd/link/internal/ld: deflake TestWindowsIssue36495 Over a dozen of the ld tests were missing closes. That was less obvious before CL 299670 started using T.TempDir instead, which fails a test when the tempdir can't be cleaned up (as it can't on Windows when things are still open), insteading of leaving tempdirs aroun...
[ { "path": "src/cmd/link/internal/ld/dwarf_test.go", "patch": "@@ -1297,6 +1297,7 @@ func main() {\n fmt.Println(\"Hello World\")\n }`\n \tf := gobuild(t, dir, prog, NoOpt)\n+\tdefer f.Close()\n \texe, err := pe.Open(f.path)\n \tif err != nil {\n \t\tt.Fatalf(\"error opening pe file: %v\", err)", "addi...
2021-03-09T02:44:27
vercel/next.js
be1537881ca0893f8a422b2cd6d59ad5ae8a4004
c5bc290c329a03b4053833f705e6ac62ad92d51c
DeterminsticHash cleanup (vercel/turbo#108) Fixes a few issues: 1. Removes `Hasher` impl for `Xxh3Hash64Hasher` - Importing the `Hash` trait allows you to call `vc.hash(&mut xxh3_hasher)`, which defeats our goal of deterministic hashing 2. Fixes hashing of 2 contiguous strings - `"fo"` and `"o"` should hash ...
[ { "path": "packages/next-swc/crates/next-core/js/tsconfig.json", "patch": "@@ -26,7 +26,7 @@\n \"@vercel/turbopack-next/*\": [\"src/*\"],\n \"@vercel/turbopack-next/pages/_app\": [\"node_modules/next/app\"],\n \"@vercel/turbopack-next/pages/_document\": [\"node_modules/next/document\"],\n-...
2022-10-19T22:21:29
ollama/ollama
c9b2dcfc52e6fe172ea1169f94f1f6839a822c09
b00bd1dfd4e9c8cf012eb8a1d2e406565f00d13e
anthropic: fix empty inputs in content blocks (#15105) * anthropic: fix empty inputs in content blocks When we switched to `api.ToolCallFunctionArguments`, `omitempty` stopped doing what we were relying on it for before. This would cause non-tool content blocks to have an `"input": {}` field, which doesn't match our ...
[ { "path": "anthropic/anthropic.go", "patch": "@@ -123,7 +123,7 @@ type ContentBlock struct {\n \t// For tool_use and server_tool_use blocks\n \tID string `json:\"id,omitempty\"`\n \tName string `json:\"name,omitempty\"`\n-\tInput api.ToolCallFunctionArgument...
2026-03-27T22:41:27
golang/go
b6def6a34e049d5d2cc9225d991c4b84427467ec
a08adda12c2af7f597dad9d280282c7e70d4a2dc
cmd/vet: bring in sigchanyzer to report unbuffered channels to signal.Notify Brings in the static analyzer "sigchanyzer", that we created at Orijtech, Inc, and already submitted in CL 274352, as golang.org/x/tools/go/analysis/passes/sigchanyzer and add it to cmd/vet as one of the passes. Fixes #9399 Change-Id:...
[ { "path": "src/cmd/vendor/golang.org/x/tools/go/analysis/passes/sigchanyzer/sigchanyzer.go", "patch": "@@ -0,0 +1,129 @@\n+// Copyright 2020 The Go Authors. All rights reserved.\n+// Use of this source code is governed by a BSD-style\n+// license that can be found in the LICENSE file.\n+\n+// Package sigcha...
2021-03-07T07:57:33
huggingface/transformers
6f127d3f81c8cf434aa809a4d8ec76b3a6372060
6b73ee890531c3f66f52a4090d64cf6348c51ad7
Skip `torchscript` tests if a cache object is in model's outputs (#35596) * fix 1 * fix 1 * comment --------- Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
[ { "path": "tests/test_modeling_common.py", "patch": "@@ -122,7 +122,7 @@\n from torch import nn\n \n from transformers import MODEL_MAPPING, AdaptiveEmbedding\n- from transformers.cache_utils import DynamicCache\n+ from transformers.cache_utils import Cache, DynamicCache\n from transformer...
2025-01-10T09:46:03
nodejs/node
f7496896c4ea088cbd5e4ed29eeecca655e1ac0b
2c3146d06dfc1815915232eae38da3cdeaaaeb13
tools: add eslintrc rule for `assert.rejects` This makes sure `assert.rejects` is always called with a second argument. Besides that it is also changes a eslint error message to suggest objects instead of a regular expression. PR-URL: https://github.com/nodejs/node/pull/19885 Reviewed-By: Matteo Collina <matteo.colli...
[ { "path": ".eslintrc.js", "patch": "@@ -155,9 +155,13 @@ module.exports = {\n selector: \"CallExpression[callee.object.name='assert'][callee.property.name='doesNotThrow']\",\n message: \"Please replace `assert.doesNotThrow()` and add a comment next to the code instead.\"\n },\n+ {...
2018-04-08T22:12:48
vercel/next.js
a3e10b566de2cf1c9a46f9b0fe9aa0b870295a2d
2a483842536fe34381bd4d7d671cad634f766397
DeterminsticHash cleanup (#108) Fixes a few issues: 1. Removes `Hasher` impl for `Xxh3Hash64Hasher` - Importing the `Hash` trait allows you to call `vc.hash(&mut xxh3_hasher)`, which defeats our goal of deterministic hashing 2. Fixes hashing of 2 contiguous strings - `"fo"` and `"o"` should hash differently ...
[ { "path": "crates/next-core/js/tsconfig.json", "patch": "@@ -26,7 +26,7 @@\n \"@vercel/turbopack-next/*\": [\"src/*\"],\n \"@vercel/turbopack-next/pages/_app\": [\"node_modules/next/app\"],\n \"@vercel/turbopack-next/pages/_document\": [\"node_modules/next/document\"],\n- \"next/*\": ...
2022-10-19T22:21:29
rust-lang/rust
f6faaee372c1f7fcf2b9ab49310ed07a94620722
f06e5c1e35bc5bc6131c6f8a0eb782097e3f28c3
Report higher-ranked trait error when higher-ranked projection goal fails in new solver
[ { "path": "compiler/rustc_trait_selection/src/solve/fulfill/derive_errors.rs", "patch": "@@ -291,6 +291,34 @@ impl<'tcx> BestObligation<'tcx> {\n }\n }\n \n+ /// When a higher-ranked projection goal fails, check that the corresponding\n+ /// higher-ranked trait goal holds or not. This is b...
2025-04-08T05:18:17
ollama/ollama
ac83ac20c444656f0f7d5bbad5b62da389395439
e7ccc129ea45cd9383d91f0c233f324a95ad0572
anthropic: fix KV cache reuse degraded by tool call argument reordering Use typed structs for tool call arguments instead of map[string]any to preserve JSON key order, which Go maps do not guarantee.
[ { "path": "anthropic/anthropic.go", "patch": "@@ -68,7 +68,7 @@ type MessagesRequest struct {\n \tModel string `json:\"model\"`\n \tMaxTokens int `json:\"max_tokens\"`\n \tMessages []MessageParam `json:\"messages\"`\n-\tSystem any `json:\"system,omit...
2026-03-09T23:24:57
huggingface/transformers
1e3ddcb2d0380d0d909a44edc217dff68956ec5e
e97d7a5be545d123862d09f6c32e8d90b98bbf0c
ModernBERT bug fixes (#35404) * bug fixes * organize imports * wrap cpu warning in reference_compile * Avoid needing repad_logits_with_grad, always repad with grads when training I'm not 100% that the conditional with "or labels is None" makes sense though - not sure what the intention is there. Perhaps we...
[ { "path": "docs/source/en/_toctree.yml", "patch": "@@ -505,7 +505,7 @@\n - local: model_doc/mobilebert\n title: MobileBERT\n - local: model_doc/modernbert\n- title: ModernBert\n+ title: ModernBERT\n - local: model_doc/mpnet\n title: MPNet\n - local: mode...
2025-01-09T19:15:38
golang/go
a08adda12c2af7f597dad9d280282c7e70d4a2dc
bd372847849c9187fd6112bd3cc0203c15c3ac72
os/signal: remove comments about SA_RESTART It's no longer necessary for non-Go signal handlers to use SA_RESTART. For #20400 Fixes #44761 Change-Id: Ie3c7fecc631a4a2822331754296ea09b308e1391 Reviewed-on: https://go-review.googlesource.com/c/go/+/298269 Trust: Ian Lance Taylor <iant@golang.org> Run-TryBot: Ian Lance...
[ { "path": "src/os/signal/doc.go", "patch": "@@ -129,9 +129,7 @@ If the non-Go code installs any signal handlers, it must use the\n SA_ONSTACK flag with sigaction. Failing to do so is likely to cause\n the program to crash if the signal is received. Go programs routinely\n run with a limited stack, and there...
2021-03-03T16:04:40
nodejs/node
85e1819d8bc8fad398ac0674064a19a691ea0fd7
c974f1bbe835f66218a15456ca9f8faee6b85c91
lib: fix coverage reporting Taking the source code of a function and running it in another context does not play well with coverage instrumentation. For now, do the simple thing and just write the source code as a string literal. Fixes: https://github.com/nodejs/node/issues/19912 Refs: https://github.com/nodejs/node/...
[ { "path": "lib/internal/util.js", "patch": "@@ -390,17 +390,16 @@ function isInsideNodeModules() {\n // Use `runInNewContext()` to get something tamper-proof and\n // side-effect-free. Since this is currently only used for a deprecated API,\n // the perf implications should be okay.\n- getStr...
2018-04-14T17:01:37
vercel/next.js
008a962df82fbe9558f95fd5c5b1cbbf1fb974e0
d76d8270d60bf732a09a273095a7304b7511d9e2
Run all dev tests with replay on retry (#41168) <!-- Thanks for opening a PR! Your contribution is much appreciated. To make sure your PR is handled as smoothly as possible we request that you follow the checklist sections below. Choose the right checklist for the change that you're making: --> ## Bug - [ ]...
[ { "path": ".github/workflows/build_test_deploy.yml", "patch": "@@ -286,9 +286,25 @@ jobs:\n timeout-minutes: 10\n if: ${{needs.build.outputs.docsChange == 'nope'}}\n \n+ - run: npx @replayio/playwright install chromium\n+ if: ${{needs.build.outputs.docsChange == 'nope'}}\n+\n ...
2022-10-19T22:15:36
huggingface/transformers
e97d7a5be545d123862d09f6c32e8d90b98bbf0c
c9c682d19c9e92a261c3879322016e61095e37cb
add `_supports_flex_attn = True` for models that do support it (#35598) * add `_supports_flex_attn = True` * fix repo consistency
[ { "path": "src/transformers/models/aria/modeling_aria.py", "patch": "@@ -704,6 +704,7 @@ class AriaPreTrainedModel(PreTrainedModel):\n _skip_keys_device_placement = [\"past_key_values\"]\n _supports_flash_attn_2 = True\n _supports_sdpa = True\n+ _supports_flex_attn = True\n _supports_cach...
2025-01-09T19:03:33
ollama/ollama
e7ccc129ea45cd9383d91f0c233f324a95ad0572
69ed0c2729cfc9d30c797bd34bf12f5c6baaf373
app: fix false "out of date" model warnings (#15101) The staleness check compared the local manifest digest (SHA256 of the file on disk) against the registry's Ollama-Content-Digest header. These never matched because PullModel re-serializes the manifest JSON before writing, producing different bytes than the registry...
[ { "path": "app/ui/app/codegen/gotypes.gen.ts", "patch": "@@ -550,14 +550,12 @@ export class Error {\n }\n }\n export class ModelUpstreamResponse {\n- digest?: string;\n- pushTime: number;\n+ stale: boolean;\n error?: string;\n \n constructor(source: any = {}) {\n if ('string' ==...
2026-03-27T21:15:10
rust-lang/rust
63dcac8423750369ee251e2bad0945ad765eaf5c
b06a88ff61acb08dd4f7502908485a3eaa74d0a6
skip `tests/codegen/swap-small-types` when debug assertions are on In `swap_nonoverlapping_short` there's a new `debug_assert!`, and if that's enabled then the `alloca`s don't optimize out.
[ { "path": "tests/codegen/swap-small-types.rs", "patch": "@@ -1,6 +1,7 @@\n //@ compile-flags: -Copt-level=3 -Z merge-functions=disabled\n //@ only-x86_64\n //@ min-llvm-version: 20\n+//@ ignore-std-debug-assertions (`ptr::swap_nonoverlapping` has one which blocks some optimizations)\n \n #![crate_type = \"l...
2025-04-09T17:44:49
golang/go
fee3cd4250843a0a7c056fed3d3e6e1a423f3120
7419a86c824efc6e2696f29e4dc1ac81756f1dfb
cmd/compile: fix width not calculated for imported type The compiler currently has problem that some imported type is missing size calculation. The problem is not triggered until CL 283313 merged, due to the compiler can compile the functions immediately when it sees them, so during SSA generation, size calculation is...
[ { "path": "src/cmd/compile/internal/typecheck/iimport.go", "patch": "@@ -508,6 +508,12 @@ func (p *iimporter) typAt(off uint64) *types.Type {\n \t\t\tbase.Fatalf(\"predeclared type missing from cache: %d\", off)\n \t\t}\n \t\tt = p.newReader(off-predeclReserved, nil).typ1()\n+\t\t// Ensure size is calculate...
2021-03-08T18:52:10
nodejs/node
cc7bef9eed6641aa0e821139b9d5227f4ae81496
e76831be0ca92585f2e4a3caf26875c6963b6019
errors: alter ERR_HTTP2_PSEUDOHEADER_NOT_ALLOWED changes the base instance for ERR_HTTP2_PSEUDOHEADER_NOT_ALLOWED from Error to TypeError as a more accurate representation of the error.. PR-URL: https://github.com/nodejs/node/pull/19958 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Yuta Hiroto <hello@hi...
[ { "path": "lib/internal/errors.js", "patch": "@@ -826,10 +826,8 @@ E('ERR_HTTP2_PAYLOAD_FORBIDDEN',\n 'Responses with %s status must not have a payload', Error);\n E('ERR_HTTP2_PING_CANCEL', 'HTTP2 ping cancelled', Error);\n E('ERR_HTTP2_PING_LENGTH', 'HTTP2 ping payload must be 8 bytes', RangeError);\n-\...
2018-04-04T19:11:27
huggingface/transformers
3a4ae6eace078c6d3c0f064b246cf9bde8978812
32e0db8a693d32963e4a0da83bc3ad87bc820835
Refactor/fix Cohere2 (#35594) * refactor/fix cohere2 * add kwargs * tests * remove func and import it
[ { "path": "src/transformers/models/cohere2/modeling_cohere2.py", "patch": "@@ -19,8 +19,7 @@\n # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n # See the License for the specific language governing permissions and\n # limitations under the License.\n-import math\n-from typing imp...
2025-01-09T16:54:57
vercel/next.js
d76d8270d60bf732a09a273095a7304b7511d9e2
f055b167a5d8e57bd9b85613f4d4ea470dbaf4b7
Update RSC detection in minimal mode and fix config collection (#41541) Updates our RSC detection for revalidation in minimal mode and also ensures we await import promises when collecting app dir config. x-ref: [slack thread](https://vercel.slack.com/archives/C035J346QQL/p1666202250144319) x-ref: [slack thread](htt...
[ { "path": "packages/next/build/utils.ts", "patch": "@@ -1051,14 +1051,14 @@ type GenerateParams = Array<{\n isLayout?: boolean\n }>\n \n-export const collectGenerateParams = (\n+export const collectGenerateParams = async (\n segment: any,\n parentSegments: string[] = [],\n generateParams: GeneratePa...
2022-10-19T21:38:54
ollama/ollama
d1151e18a1a689e47d2c1c27538213ae412d0297
ebbce136c7e035872ca7027dc9438bc2efc17123
mlx: fix KV cache snapshot memory leak mlx.Copy shares the backing buffer with its source (via copy_shared_buffer) rather than allocating independent storage. When used to snapshot a slice of the KV cache, the snapshot array holds the entire original cache buffer alive through the shared data pointer — even after eval...
[ { "path": "x/mlxrunner/cache/cache.go", "patch": "@@ -109,8 +109,8 @@ func (c *KVCache) Snapshot(fromOffset int) Snapshot {\n \n \tkSlice := c.keys.Slice(mlx.Slice(), mlx.Slice(), mlx.Slice(from, to), mlx.Slice())\n \tvSlice := c.values.Slice(mlx.Slice(), mlx.Slice(), mlx.Slice(from, to), mlx.Slice())\n-\tk...
2026-03-25T17:47:59
golang/go
7419a86c824efc6e2696f29e4dc1ac81756f1dfb
b4787201c9872fbbe8e1ebb2e2113b31ec2105a6
cmd/link/internal/ld: fix typo in a comment Change-Id: I6d0fafd38f0ad9392f163a9d8cd94d103bf2a059 Reviewed-on: https://go-review.googlesource.com/c/go/+/299669 Reviewed-by: Keith Randall <khr@golang.org> Reviewed-by: Cherry Zhang <cherryyz@google.com>
[ { "path": "src/cmd/link/internal/ld/deadcode.go", "patch": "@@ -19,7 +19,7 @@ var _ = fmt.Print\n type deadcodePass struct {\n \tctxt *Link\n \tldr *loader.Loader\n-\twq heap // work queue, using min-heap for beter locality\n+\twq heap // work queue, using min-heap for better locality\n \n \tifaceMetho...
2021-03-08T04:32:38
nodejs/node
e76831be0ca92585f2e4a3caf26875c6963b6019
45f6dccae1c98ef50d573786b25720ff0fcff5aa
stream: fix incorrect comment in _stream_readable.js PR-URL: https://github.com/nodejs/node/pull/19882 Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com>
[ { "path": "lib/_stream_readable.js", "patch": "@@ -302,7 +302,7 @@ function chunkInvalid(state, chunk) {\n }\n \n \n-// if it's past the high water mark, we can push in some more.\n+// We can push more data if we are below the highWaterMark.\n // Also, if we have no data yet, we can stand some\n // more byt...
2018-04-08T17:33:00
huggingface/transformers
32e0db8a693d32963e4a0da83bc3ad87bc820835
46276f9a7f99a9380ec86f7ecad14b3818b78a54
[`tokenizers`] Ensure that add_prefix_space is propagated to backend_tokenizer.pre_tokenizer (#35593) * Ensure that add_prefix_space is propagated to backend_tokenizer.pre_tokenizer in PreTrainedTokenizerFast, rather than relying on subclasses to take care of this. * Simplify setting self.add_prefix_space, ensur...
[ { "path": "src/transformers/models/bart/tokenization_bart_fast.py", "patch": "@@ -16,7 +16,7 @@\n import json\n from typing import List, Optional, Tuple\n \n-from tokenizers import pre_tokenizers, processors\n+from tokenizers import processors\n \n from ...tokenization_utils_base import AddedToken, BatchEnc...
2025-01-09T16:46:50
vercel/next.js
f055b167a5d8e57bd9b85613f4d4ea470dbaf4b7
fe762d329fdf9d6d6e6bf084198ad49f696d42b7
Add TypeScript plugin (#41569) This PR implements the TypeScript plugin (only applied to the app directory), and it will be automatically added to tsconfig.json when running the CLI. Due to how TS plugins work (a package name is required to specify), hence we can't use a subpath import like `next/typescript` for the p...
[ { "path": "packages/next/lib/typescript/writeConfigurationDefaults.ts", "patch": "@@ -187,6 +187,27 @@ export async function writeConfigurationDefaults(\n )\n }\n \n+ // Enable the Next.js typescript plugin.\n+ if (isAppDirEnabled) {\n+ if (userTsConfig.compilerOptions) {\n+ if (!('plugins' ...
2022-10-19T20:59:39
ollama/ollama
4fda69809a3fecf73d0f71657ef50f8f7e8f43f7
c9b5da6b0c862ccf6c92f89a26aad27442d4c59b
ci: fix windows cgo compiler error (#15046)
[ { "path": ".github/workflows/test.yaml", "patch": "@@ -64,6 +64,7 @@ jobs:\n container: nvidia/cuda:13.0.0-devel-ubuntu22.04\n extra-packages: libcudnn9-dev-cuda-13 libopenblas-dev liblapack-dev liblapacke-dev git curl\n flags: '-DCMAKE_CUDA_ARCHITECTURES=87 -DBLAS_INCLUD...
2026-03-24T23:45:36
golang/go
aafad20b617ee63d58fcd4f6e0d98fe27760678c
125eca0f7210da1bbf1a4a1460a87d1c33366b99
encoding/binary: limit bytes read by Uvarint to <= 10 Limits the number of bytes that can be consumed by Uvarint to MaxVarintLen64 (10) to avoid wasted computations. With this change, if Uvarint reads more than MaxVarintLen64 bytes, it'll return the erroring byte count of n=-(MaxVarintLen64+1) which is -11, as per the...
[ { "path": "src/encoding/binary/varint.go", "patch": "@@ -61,8 +61,13 @@ func Uvarint(buf []byte) (uint64, int) {\n \tvar x uint64\n \tvar s uint\n \tfor i, b := range buf {\n+\t\tif i == MaxVarintLen64 {\n+\t\t\t// Catch byte reads past MaxVarintLen64.\n+\t\t\t// See issue https://golang.org/issues/41185\n+...
2021-03-07T07:14:21
nodejs/node
45f6dccae1c98ef50d573786b25720ff0fcff5aa
09c63460ebda30c7d5e7f40532a311fb2139803a
deps: cherry-pick b767cde1e7 from upstream V8 Original commit message: [intl] unbreak build with ICU 57 Remove a call to `icu::toUCharPtr()` that wasn't present in other similar looking call sites either, just reinterpret_cast directly. Fixes https://github.com/nodejs/node/issues/19656. Cq-Incl...
[ { "path": "common.gypi", "patch": "@@ -27,7 +27,7 @@\n \n # Reset this number to 0 on major V8 upgrades.\n # Increment by one for each non-official patch applied to deps/v8.\n- 'v8_embedder_string': '-node.2',\n+ 'v8_embedder_string': '-node.3',\n \n # Enable disassembler for `--print-code...
2018-03-30T10:47:02
huggingface/transformers
46276f9a7f99a9380ec86f7ecad14b3818b78a54
d3fe9fa3fec08d8793f9ba7bab991a2908393304
Fix modular edge case + modular sorting order (#35562) * look-ahead negation * re add examples by default * Fix the bug in topological sort * Update create_dependency_mapping.py * start adding test * finalize test * more tests * style * style
[ { "path": "examples/modular-transformers/configuration_my_new_model.py", "patch": "@@ -43,7 +43,7 @@ class MyNewModelConfig(PretrainedConfig):\n The non-linear activation function (function or string) in the decoder.\n max_position_embeddings (`int`, *optional*, defaults to 2048):\n ...
2025-01-09T16:17:52
ollama/ollama
95ee7fbd29c6a0ceff073db87b81b005cd15b9c5
ec55536734976ae12f0186b667cd698d1e9f92d2
mlxrunner: panic on double unpin
[ { "path": "x/mlxrunner/cache/recurrent.go", "patch": "@@ -22,14 +22,9 @@ func (c *RecurrentCache) setStateRaw(old, v *mlx.Array) *mlx.Array {\n \tif v == nil || !v.Valid() {\n \t\treturn old\n \t}\n-\tif old == v {\n-\t\treturn old\n-\t}\n \n \tmlx.Pin(v)\n-\tif old != nil && old != v {\n-\t\tmlx.Unpin(old)...
2026-03-20T23:10:19