JamieBradfield/qwen3.8-9b-hermes-fc-todo-GGUF overview
Qwen3.8 9B Hermes FC — Todo GGUF ROCmFPX quantized version of JamieBradfield/qwen3.8 9b hermes fc todo https://huggingface.co/JamieBradfield/qwen3.8 9b hermes …
Runs locally from ~4.58 GB disk (8 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| qwen3.8-9b-hf-fc-v27-Q4_0_ROCMFP4_FAST.gguf | GGUF | Q4_0_ROCMFP4_FAST | 4.58 GB | Download |
Model Details
| Model ID | JamieBradfield/qwen3.8-9b-hermes-fc-todo-GGUF |
|---|---|
| Author | JamieBradfield |
| Pipeline | text-generation |
| License | apache-2.0 |
| Base model | Empero/Qwen3.8-9B |
| Last modified | 2026-08-31T18:00:24.000Z |
Model README
---
license: apache-2.0
base_model: Empero/Qwen3.8-9B
tags:
- qwen3.5
- function-calling
- tool-use
- todo
- qlora
- finetune
- gguf
- rocmfpx
pipeline_tag: text-generation
language:
- en
library_name: gguf
---
Qwen3.8-9B Hermes FC — Todo (GGUF)
ROCmFPX-quantized version of
JamieBradfield/qwen3.8-9b-hermes-fc-todo
(BF16 merge in the parent repo).
Quant details
| file | quant | size | notes |
|---|---|---|---|
| qwen3.8-9b-hf-fc-v27-Q4_0_ROCMFP4_FAST.gguf | Q4_0_ROCMFP4_FAST | 4.69 GB | ROCmFPX (AMD RDNA3 kernels); fast-path quant of the BF16 merge |
Convert/quantize: llama-rocmfpx fork, convert_hf_to_gguf.py --outtype bf16
→ llama-quantize Q4_0_ROCMFP4_FAST.
ROCmFPX quants target AMD ROCm inference (RX 7700 XT in the author's rig,
12 GB VRAM, served at 245k context with q8_0/turbo3 KV). For portable use,
convert from the BF16 merge in the parent repo instead.
Evaluation
See the parent repo card for the full 40-probe battery. Headline: tier-1
held-out 13/20, tier-2 autonomous todo reach-for 7/10 (up from 3/10 on v26).
Run JamieBradfield/qwen3.8-9b-hermes-fc-todo-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models