What's Changed
- Move linear spec handling to model backends by @Blaizzy in #1259
- Add NVIDIA LocateAnything-3B (MoonViT + Qwen2.5, AR + Parallel Box Decoding) by @beshkenadze in #1242
- Fix Gemma 4 MTP rollback crash when accepted is a list (#1260) by @francip in #1261
- [Cohere] Add cohere2_moe model support by @Terrencezzj in #1268
- Fix system role normalization for /v1/messages API by @lucasnewman in #1269
- Add Gemma 4 Unified + MTP support by @Blaizzy in #1267
- Fix Gemma 4 rollback handling and streaming thinking splits by @Blaizzy in #1266
New Contributors
- @francip made their first contribution in #1261
- @Terrencezzj made their first contribution in #1268
Full Changelog: v0.6.0...v0.6.1