Skip to content

Pull requests: Luce-Org/lucebox-hub

New pull request New
33 Open 191 Closed
33 Open 191 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

fix(server): Qwen3.6-27B tool calling for claude-code Anthropic path
#276 opened May 25, 2026 by dusterbloom Contributor Loading…
5 of 7 tasks
feat(drafter): ee3 as production default (depends on #274)
#275 opened May 24, 2026 by dusterbloom Contributor Loading…
6
feat(pFlash): ee7 early-exit drafter saves up to 9.3× drafter wall at 128K
#274 opened May 24, 2026 by dusterbloom Contributor Loading…
28
feat(server): add Gemma4 target-layer-split adapter
#273 opened May 24, 2026 by weicj Contributor • Draft
feat(cpp-server): thinking-budget v2 + multi-dialect reasoning aliases + spec
#269 opened May 23, 2026 by easel Contributor Loading…
6 tasks done
3
feat(harness): typed adapters + format-aware session-inject proxy + multi-turn bandit driver
#266 opened May 23, 2026 by dusterbloom Contributor Loading…
2
feat(server): add target-layer-split backend adapter path
#265 opened May 23, 2026 by weicj Contributor • Draft
3
feat(pflash): adaptive keep_ratio bandit MVP
#264 opened May 23, 2026 by dusterbloom Contributor Loading…
9
feat(server): add mixed-backend PFlash phase split
#263 opened May 23, 2026 by weicj Contributor Loading…
1
Split MoE weights between CPU & CUDA, support qwen35moe models
#262 opened May 23, 2026 by howard0su Contributor Loading…
1 of 3 tasks
4
feat(pflash): add dynamic threshold for chunk selection and lower auto-mode trigger
#249 opened May 22, 2026 by howard0su Contributor • Draft
feat(dFlash): MTP foundation
#237 opened May 20, 2026 by dusterbloom Contributor Loading…
25
docker, cli, smoke, bench and autotune first-run dx
#226 opened May 19, 2026 by easel Contributor • Draft
16
mtp: prefix-cache WARM hit (perfect + partial via range-warm)
#221 opened May 18, 2026 by dusterbloom Contributor Loading…
7
feat(gemma4): feature-complete backend with DFlash + MTP + sparse-FA decode (supersedes PR #175 skeleton)
#193 opened May 14, 2026 by dusterbloom Contributor • Draft
feat(gemma4): target-graph MTP integration (h_prev capture + asymmetric KV)
#183 opened May 13, 2026 by dusterbloom Contributor Loading…
5
feat(gemma4): add mtp loader and step graph
#182 opened May 13, 2026 by dusterbloom Contributor Loading…
6
feat(gemma4): add dflash draft runtime
#181 opened May 13, 2026 by dusterbloom Contributor Loading…
5
feat(gemma4): add draft loader and quantization support
#180 opened May 13, 2026 by dusterbloom Contributor Loading…
9
fix(gemma4): add long-context KV correctness
#177 opened May 13, 2026 by dusterbloom Contributor Loading…
5
docs(dflash): document small-vram cuda vmm guidance
#174 opened May 13, 2026 by dusterbloom Contributor Loading…
4
feat(dflash): linear native MTP integrated decode CLI (stacked on #153)
#154 opened May 11, 2026 by javierpazo Contributor Loading…
2
feat(dflash): native Qwen3.6 MTP (NextN) runtime + contract test
#153 opened May 11, 2026 by javierpazo Contributor Loading…
3
Add Gemma 4 RTX 4090 backend helpers
#152 opened May 11, 2026 by adybag14-cyber Loading…
@dusterbloom
14
feat(dflash): accept FP16 safetensors drafter alongside BF16
#142 opened May 9, 2026 by javierpazo Contributor Loading…
ProTip! no:milestone will show everything without a milestone.