---

license: apache-2.0

base_model: InternScience/Agents-A1

library_name: llama.cpp

tags:

• gguf
• llama.cpp
• imatrix
• fable5-calibrated
• agents-a1

---

Agents-A1-Q2_K-imatrix-gguf-fable5-calibrated

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

File

| File | Size | SHA-256 |

|---|---:|---|

| Agents-A1-Q2_K-imatrix-gguf-fable5-calibrated.gguf | 12.05 GiB | 58d9f6e1de32a960c4dca0dca932241956303687b3a7436f8852a69812864067 |

Quick Start

llama-cli -hf Chungulus/Agents-A1-Q2_K-imatrix-gguf-fable5-calibrated:Agents-A1-Q2_K-imatrix-gguf-fable5-calibrated.gguf -p "Write a Python sorting function" -n 160

Ollama

ollama create agents-a1-q2_k-imatrix -f Modelfile
ollama run agents-a1-q2_k-imatrix

Which File Should I Download?

| Use case | Recommendation |

|---|---|

| Recommended hardware | 8-12 GB RAM |

| Best for | smallest generated file |

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 | 0.128242 |

| Same top p | 81.75% |

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.