README.md

---
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
<function name="memory_add">
  <param name="text">Remember that the brain slot uses the ROCmFPX runner.</param>
</function>
```

The hal0 integration validates the tool schema, allows at most one complete tool call per turn, and stops after `</function>`.

## 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)