Skip to content

decision2:lux

6 TagsUpdated 7.94B params16384 contextEnglishApache-2.0by vLLM Semantic Router

Qwen3.5-9B, the largest: 0.707 on typed decisions.

7.94b
ollaya run decision2:lux --preset triage "I was charged twice for my subscription this month and want a refund."

Details

  • graph8046e5224728 · 12 MBonnx · decision2 · 7.94B · fp32
  • weights03cd850d6640 · 4.1 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00001-of-00009.safetensors
  • weightscad4b1e5f7ca · 2.0 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00002-of-00009.safetensors
  • weights6ac88bf9ed3b · 2.0 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00003-of-00009.safetensors
  • weights0e2f45de6cc1 · 2.0 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00004-of-00009.safetensors
  • weights214acdbdf538 · 2.0 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00005-of-00009.safetensors
  • weights05b290c54283 · 1.9 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00006-of-00009.safetensors
  • weights5e934511ffe2 · 1.9 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00007-of-00009.safetensors
  • weights17476a392385 · 2.0 GBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00008-of-00009.safetensors
  • weights30952702ff21 · 76 MBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/backbone/model-00009-of-00009.safetensors
  • weights47f7b4d7e781 · 17 MBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/decision_head.safetensors
  • tokenizer06b9509352d2 · 20 MBhuggingface.co/vllm-sr/Decision-2.0-Lux-9B/resolve/1dcd1e7…/tokenizer.json
  • decision010c4a59986d · 5 KB{"engine": "onnx", "family": "decision2", "encoder": "", "layout": "decision2-endpoint-v1", …}
  • calibrationbf756c46be45 · 187 B{"temperature": [1.0, 1.0, 1.0]}
  • license1314ad54d94f · 10 KBDecision 2.0 by the vLLM Semantic Router contributors (https://huggingface.co/vllm-sr)

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.

Decision 2.0 readme and all models