docs(research): add qwen3.8-27b decode speed analysis #68

Merged
haylan merged 1 commits from docs/decode-speed-research into main 2026-11-11 15:52:21 +00:00
Owner
  • Add research note on decode tokens/s versus context depth for Qwen3.8-27B + DFlash2 on the R9700
  • Covers draft length (5, not 7), acceptance falling with depth, bandwidth ceiling, and an ordered experiment plan. Proposals only, nothing applied
  • Index the note in docs/research/README.md
  • Also stages .serena/ project config and llmserver.log
- Add research note on decode tokens/s versus context depth for Qwen3.8-27B + DFlash2 on the R9700 - Covers draft length (5, not 7), acceptance falling with depth, bandwidth ceiling, and an ordered experiment plan. Proposals only, nothing applied - Index the note in docs/research/README.md - Also stages .serena/ project config and llmserver.log
haylan added 1 commit 2026-11-11 15:52:17 +00:00
- Add research note on decode tokens/s versus context depth for Qwen3.8-27B + DFlash2 on the R9700
- Covers draft length (5, not 7), acceptance falling with depth, bandwidth ceiling, and an ordered experiment plan. Proposals only, nothing applied
- Index the note in docs/research/README.md
- Also stages .serena/ project config and llmserver.log
haylan merged commit 71fb83f760 into main 2026-11-11 15:52:21 +00:00
haylan deleted branch docs/decode-speed-research 2026-11-11 15:52:22 +00:00
Sign in to join this conversation.
No Reviewers
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: haylan/LLM-Server#68