---
license: apache-2.0
base_model: InternScience/Agents-A1
library_name: llama.cpp
tags:
- gguf
- llama.cpp
- imatrix
- fable5-calibrated
- agents-a1
---

# Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated

Fable-5 trace calibrated imatrix GGUF quant of `InternScience/Agents-A1`.

## File

| File | Size | SHA-256 |
|---|---:|---|
| `Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated.gguf` | 19.71 GiB | `06361f183ec008c7052c0473a746f867c25779b1debb4a8a74a7cee27abc33d2` |

## Quick Start

```bash
llama-cli -hf Chungulus/Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated:Agents-A1-Q4_K_M-imatrix-gguf-fable5-calibrated.gguf -p "Write a Python sorting function" -n 160
```

## Ollama

```bash
ollama create agents-a1-q4_k_m-imatrix -f Modelfile
ollama run agents-a1-q4_k_m-imatrix
```

## Which File Should I Download?

| Use case | Recommendation |
|---|---|
| Recommended hardware | 16-24 GB RAM |
| Best for | default recommendation |

## Quality Snapshot

F16 baseline mini accuracy: 89.58%.
F16 baseline PPL on KL holdout: 13.0194.

| Metric | Value |
|---|---:|
| Mini accuracy | 87.50% |
| Retention vs F16 | 97.67% |
| Mean KLD vs F16 | 0.015182 |
| Same top p | 93.65% |

## Notes

- Calibration source: `Glint-Research/Fable-5-traces`
- Calibration source revision: `e05c417852fc59fd8da758e68b352732423ca0cb`
- GGUF quantization method: llama.cpp with imatrix calibration.
- Static imatrix GGUF; not Unsloth Dynamic 2.0 / UD2.
- MTP is not included because the downloaded checkpoint did not contain MTP tensors.
- This repo contains local quantization artifacts only.