Feb 24

1. 20:53

   docs: show estimated quality and cost in example output

   JiaqiXue
   Verified signature

   8de7b55
2. 20:50

   docs: use generic predictor terminology in training output

   JiaqiXue
   Verified signature

   3858b5d
3. 20:44

   feat: add training progress prints to from_training_data()

   JiaqiXue
   Verified signature

   9e08fe1
4. 20:44

   docs: simplify README, remove GPU/CPU examples

   JiaqiXue
   Verified signature

   8ac636e
5. 20:35

   docs: add lambda parameter documentation with examples

   JiaqiXue
   Verified signature

   47a8f8e
6. 20:21

   fix: update vllm serve comment to --runner pooling

   JiaqiXue
   Verified signature

   603d970
7. 20:21

   fix: use --runner pooling instead of --task embed for vllm serve

   JiaqiXue
   Verified signature

   82903a0
8. 20:09

   Fix vllm serve command for different versions

   JiaqiXue
   Verified signature

   8c269dd
9. 20:03

   Add vLLM server mode: start once, route from anywhere

   JiaqiXue
   Verified signature

   52e5f26
10. 19:55

    vLLM-based route_text(), uv install guide, complete runnable examples

    JiaqiXue
    Verified signature

    5309389
11. 19:51

    Fix: use transformers AutoModel instead of sentence_transformers for Qwen3 compatibility

    JiaqiXue
    Verified signature

    230798d
12. 19:48

    Fix Quick Start: complete runnable example with embedding generation

    JiaqiXue
    Verified signature

    5bc5cb2
13. 19:41

    Fix Quick Start: use snapshot_download for proper import

    JiaqiXue
    Verified signature

    a43b342
14. 19:28

    Initial release: R2-Router (#1 on RouterArena)

    JiaqiXue
    Verified signature

    bc1c255
15. 19:28

    initial commit

    JiaqiXue
    Verified signature

    db29071