Chungulus/Agents-A1-Q5_K_M-imatrix-gguf-fable5-calibrated overview
A1 Q5 K M imatrix Static imatrix calibrated GGUF quant of InternScience/Agents A1 . bash llama cli hf Chungulus/Agents A1 Q5 K M imatrix gguf fable5 calibrated…
Runs locally from ~23.03 GB disk (24 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| Agents-A1-Q5_K_M-imatrix-gguf-fable5-calibrated.gguf | GGUF | Q5_K_M | 23.03 GB | Download |
Model Details
Model README
---
license: apache-2.0
base_model: InternScience/Agents-A1
library_name: llama.cpp
tags:
- gguf
- llama.cpp
- imatrix
- agents-a1
---
A1-Q5_K_M-imatrix
Static imatrix-calibrated GGUF quant of InternScience/Agents-A1.
llama-cli -hf Chungulus/Agents-A1-Q5_K_M-imatrix-gguf-fable5-calibrated:Agents-A1-Q5_K_M-imatrix-gguf-fable5-calibrated.gguf -p "Write a Python sorting function" -n 160File
| File | Size | SHA-256 |
|---|---:|---|
| Agents-A1-Q5_K_M-imatrix-gguf-fable5-calibrated.gguf | 23.03 GiB | 1e95ead178c1cfc187fcac4627515644c0105d3e5ec08f3f2fb9e317470b7840 |
Quality Snapshot
F16 baseline mini accuracy: 89.58%.
F16 baseline PPL on KL holdout: 13.0194.
| Metric | Value |
|---|---:|
| Mini accuracy | 89.58% |
| Retention vs F16 | 100.00% |
| Mean KLD vs F16 | 0.007785 |
| Same top p | 96.83% |
Notes
- Calibration source:
Glint-Research/Fable-5-traces - MTP is not included because the downloaded checkpoint did not contain MTP tensors.
- Static imatrix GGUF, not Unsloth Dynamic 2.0 / UD2.
Run Chungulus/Agents-A1-Q5_K_M-imatrix-gguf-fable5-calibrated with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models