--- license: apache-2.0 pipeline_tag: text-generation base_model: Hal0ai/hal0-brain-sft base_model_relation: quantized tags: - hal0 - hal0-brain - gguf - rocmfpx - rocmfp4 - rocmfp8 - agent - tool-use language: - en --- ![HAL0 BRAIN — advanced reasoning, tool calling, and platform management](hal0-brain-banner.png) # HAL0 BRAIN — ROCmFPX GGUF **Advanced reasoning · Tool calling · Platform management** HAL0 BRAIN is a mini-agent administrator trained on the hal0 system: its services, profiles, slots, model lifecycle, diagnostics, and tool workflows. This repository packages the verified F16 reference plus two agent-oriented ROCmFPX quants for structured reasoning, tool calling, and platform management. ## Files | File | Preset | Bytes | SHA-256 | `general.file_type` | Observed tensor types | |---|---:|---:|---|---:|---| | `hal0-brain-sft-F16.gguf` | F16 reference | 2,166,552,096 | `ed9d28c4eac1d7c291bc80d9410c243a3d28e655921ccaf90f2b6619aa24d2c3` | 1 | 170 `F16`, 49 `F32` | | `hal0-brain-sft-Q4_0_ROCMFP4_COHERENT.gguf` | `Q4_0_ROCMFP4_COHERENT` | 664,952,352 | `062219355646f09e184cfb30cbfdfc429d70275f3be4d996d32644ec2b20db44` | 102 | 169 `Q4_0_ROCMFP4`, 1 `Q6_K`, 49 `F32` | | `hal0-brain-sft-Q8_0_ROCMFPX_AGENT.gguf` | `Q8_0_ROCMFPX_AGENT` | 1,144,682,016 | `eeb38e45877d26a5923d2ec22e1c887a0b12f0c40f9df72015bd73f11d1263e4` | 115 | 58 `Q8_0_ROCMFPX`, 112 protected `Q8_0`, 49 `F32` | | `chat-long-context.hal0profile.json` | portable hal0 profile | — | profile checksum `sha256:241af4cd2636ac1da32a8a7ca0d856724445242cfcde88a208b702b155bdee47` | — | `hal0.profile` schema 1 | The Agent presets deliberately retain higher-precision tensors where structured behavior is most sensitive. The filenames are descriptive, but the format claims above come from ROCmFPX-aware GGUF inspection—not filename inference. ## Provenance - Base model: [`Hal0ai/hal0-brain-sft`](https://huggingface.co/Hal0ai/hal0-brain-sft) - Verified F16 GGUF source: [`Hal0ai/hal0-brain-sft-GGUF`](https://huggingface.co/Hal0ai/hal0-brain-sft-GGUF) at revision `6b190df6e816cc806f7fa7ae3de7248f5551e00b` - F16 source SHA-256: `ed9d28c4eac1d7c291bc80d9410c243a3d28e655921ccaf90f2b6619aa24d2c3` - Quantizer: [`charlie12345/ROCmFPX`](https://github.com/charlie12345/ROCmFPX) at commit `61f2f2d7bc4955e9bca821095ef69125837133b5` - Architecture: Llama-compatible GGUF, approximately 1.08B parameters - License: Apache-2.0 Both quantized files were produced independently from F16. Neither was requantized from another low-bit artifact. ## Runtime compatibility | Artifact | Stock llama.cpp | ROCmFPX CPU reference | ROCmFPX HIP/ROCm | ROCmFPX Vulkan | |---|---:|---:|---:|---:| | F16 | ✅ | ✅ | ✅ | ✅ | | Q4 ROCmFP4 Coherent | ❌ | ✅ | ✅ | ✅ | | Q8 ROCmFPX Agent | ❌ | ✅ | ✅ | ✅ | The Q4 and Q8 files contain custom GGML tensor types and require a **ROCmFPX-capable runner**. Current stock llama.cpp rejects the Q4 tensor type ID `100` and Q8 tensor type ID `103` as invalid/unknown. LM Studio, Ollama, or another stock llama.cpp package must not be assumed compatible. hal0 pins the unified runner by immutable digest: ```text ghcr.io/hal0ai/hal0-rocmfpx@sha256:fd6b02a720e633e402e929e19eedefff52aeec18e5de8f43e525689e523985f3 ``` That runner contains the MiniCPM5 pre-tokenizer mapping required by this model and supports both ROCm/HIP and Vulkan ROCmFPX paths. ## Download Install the Hugging Face CLI and authenticate only if your environment requires it: ```bash pip install -U huggingface_hub ``` F16: ```bash hf download Hal0ai/hal0-brain-sft-ROCmFPX-GGUF \ hal0-brain-sft-F16.gguf --local-dir . ``` ROCmFP4 Coherent Agent: ```bash hf download Hal0ai/hal0-brain-sft-ROCmFPX-GGUF \ hal0-brain-sft-Q4_0_ROCMFP4_COHERENT.gguf --local-dir . ``` ROCmFP8 Agent: ```bash hf download Hal0ai/hal0-brain-sft-ROCmFPX-GGUF \ hal0-brain-sft-Q8_0_ROCMFPX_AGENT.gguf --local-dir . ``` Portable hal0 long-context profile: ```bash hf download Hal0ai/hal0-brain-sft-ROCmFPX-GGUF \ chat-long-context.hal0profile.json --local-dir . ``` ## Run with ROCmFPX ROCm/HIP: ```bash llama-cli \ -m hal0-brain-sft-Q8_0_ROCMFPX_AGENT.gguf \ -dev ROCm0 -ngl 999 -fa on -c 8192 -b 512 -ub 512 --jinja ``` Vulkan: ```bash llama-cli \ -m hal0-brain-sft-Q4_0_ROCMFP4_COHERENT.gguf \ -dev Vulkan0 -ngl 999 -fa on -c 8192 -b 512 -ub 512 --jinja ``` OpenAI-compatible server: ```bash llama-server \ -m hal0-brain-sft-Q8_0_ROCMFPX_AGENT.gguf \ --host 0.0.0.0 --port 8080 \ -dev Vulkan0 -ngl 999 -fa on -c 8192 -b 512 -ub 512 --jinja ``` Choose `ROCm0` or `Vulkan0` according to the devices reported by the ROCmFPX runner. Long context and large batches consume substantial memory; tune them for the host. ## Tool-calling contract hal0's catalog declares the `hal0-function-xml` prompt contract. A complete tool call has this shape: ```xml Remember that the brain slot uses the ROCmFPX runner. ``` The hal0 integration validates the tool schema, allows at most one complete tool call per turn, and stops after ``. ## Portable hal0 profile `chat-long-context.hal0profile.json` is a checksummed, portable `hal0.profile` schema-1 envelope. It includes: ```text -fa on -ctk q8_0 -ctv q8_0 -b 2048 -ub 512 --parallel 1 --no-mmap --no-context-shift --poll 100 --poll-batch 1 --metrics --no-webui ``` These are performance-oriented defaults, not a hardware guarantee. Reduce context/batch settings or use another profile when the host cannot sustain them. ### Dashboard import Open **Profiles**, choose **Import**, select `chat-long-context.hal0profile.json`, review the dry-run result, and commit the import. If `chat-long-context` already exists, choose another name rather than overwriting it implicitly. ### REST API dry-run and import Dry-run validates the envelope and checksum without changing the profile catalog: ```bash PROFILE=chat-long-context.hal0profile.json jq -n --slurpfile envelope "$PROFILE" \ '{envelope:$envelope[0],name:"chat-long-context",dry_run:true}' | curl --fail-with-body -sS http://127.0.0.1:8080/api/profiles/import \ -H 'content-type: application/json' --data-binary @- ``` Commit only after the dry-run reports `valid: true`, `checksum_ok: true`, and no unwanted collision: ```bash PROFILE=chat-long-context.hal0profile.json jq -n --slurpfile envelope "$PROFILE" \ '{envelope:$envelope[0],name:"chat-long-context",dry_run:false}' | curl --fail-with-body -sS http://127.0.0.1:8080/api/profiles/import \ -H 'content-type: application/json' --data-binary @- ``` ## Validation record - ROCmFPX-aware inspection confirmed every file type and tensor count in the table. - ROCmFPX CPU reference execution loaded and benchmarked both custom artifacts. - The catalog runner image includes the required MiniCPM5 tokenizer mapping and ROCmFPX ROCm/Vulkan kernels. - Stock llama.cpp `e8e6c7af2456fd50bb62f7a2bbd642e6fb14ae77` rejected Q4 type `100` and Q8 type `103` before tensor loading. ## Related - [hal0](https://github.com/Hal0ai/hal0) - [ROCmFPX](https://github.com/charlie12345/ROCmFPX) - [Hal0ai model collection](https://huggingface.co/collections/Hal0ai/hal0-brain-6a5e1859dbae8e442c659141)