Skip to content

decima:agent

7 TagsUpdated 321M params8192 context100+ languagesApache-2.0by A. M. Madani

Decima-agent 2.3: decima-base fine-tuned for coding-agent decisions (secret and command gates, tool and model choice) and knowledge-graph judgments; states up to 8,192 tokens, options kept whole beside long questions.

multilingual321m
ollaya run decima:agent --preset triage "I was charged twice for my subscription this month and want a refund."

Details

  • graph21aa40387afd · 5 MBonnx · decima · 321M · fp32
  • weightsc86fb570cd37 · 1.2 GBhuggingface.co/amyrmahdy/decima-agent/resolve/652d90a…/pytorch/encoder/model.safetensors
  • weights98fc67d8969c · 57 MBhuggingface.co/amyrmahdy/decima-agent/resolve/652d90a…/pytorch/head.safetensors
  • tokenizerd941237f7b66 · 34 MBhuggingface.co/amyrmahdy/decima-agent/resolve/652d90a…/pytorch/encoder/tokenizer.json
  • decisione38423617476 · 3 KB{"engine": "onnx", "family": "decima", "encoder": "", "layout": "decima-late-interaction-v2", …}
  • calibration0a26d5ad7a28 · 291 B{"temperature": [0.612, 1.0, 0.612]}
  • license6e583cb7b1fd · 11 KBDecima-small, Decima-base, Decima-agent, Decima-guard and Decima-guard-small by A. M. Madani (https://huggingface.co/amyrmahdy, https://github.com/amyrmahdy/decima), Apache-2.0.

Every layer is checked against its sha256 when it is pulled. Weights and tokenizers download from the model author's Hugging Face repository at a pinned commit; Ollaya never re-hosts them.

Decima readme and all models