Commit 325b4a7
committed
chore(release): v0.10.0
Bump app version 0.9.1 -> 0.10.0.
Since v0.9.1:
- Qwen3.5/3.6 dense (9B/27B) + native MTP self-speculative decode
- MTP accept-lever primitives (sampling + correctors)
- tuning-tab controls for disk KV cache
- fix: tools-aware prefix-cache boundary (agentic compaction no longer
cold-re-prefills the system+tools head; 98s -> ~1-2s per turn)
- fix: reclaim MLX buffer pool per gemma4 chat (no more 35 -> 2.5 tok/s
decay across repeated requests)1 parent 76ff913 commit 325b4a7
4 files changed
Lines changed: 4 additions & 4 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | 1 | | |
2 | 2 | | |
3 | | - | |
| 3 | + | |
4 | 4 | | |
5 | 5 | | |
6 | 6 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | 1 | | |
2 | 2 | | |
3 | 3 | | |
4 | | - | |
| 4 | + | |
5 | 5 | | |
6 | 6 | | |
7 | 7 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | 1 | | |
2 | 2 | | |
3 | 3 | | |
4 | | - | |
| 4 | + | |
5 | 5 | | |
6 | 6 | | |
7 | 7 | | |
| |||
0 commit comments