zijie-tian/nano-vllm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

📝 docs: add chunked attention solutions guide and update doc index

Browse Source 
Add comprehensive documentation analyzing the 32K chunked offload
accuracy issues with proposed solutions covering LSE precision,
ring buffer state management, and position encoding validation.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
Zijie Tian
2026-01-20 04:48:20 +08:00
parent 4cbd451af7
commit 6180055ed8
2 changed files with 1080 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
CLAUDE.md
View File

@@ -20,7 +20,8 @@ Nano-vLLM is a lightweight vLLM implementation (~1,200 lines) for fast offline L
| [`docs/optimization_guide.md`](docs/optimization_guide.md) | Performance optimizations: sgDMA (15x), Triton merge (4.3x), N-way pipeline (2x) |
| [`docs/known_issues.md`](docs/known_issues.md) | Documented bugs and fixes: partial last block bug, block size 4096 race condition |
| [`docs/ruler_benchmark_results_32k.md`](docs/ruler_benchmark_results_32k.md) | RULER benchmark results (32K context): 13 tasks, 92.3% accuracy, CPU offload performance |
| [`docs/ruler_32k_chunked_offload_issue.md`](docs/ruler_32k_chunked_offload_issue.md) | ⚠️ OPEN ISSUE: 32K chunked offload accuracy problem (35% error rate in RULER) |
| [`docs/ruler_32k_chunked_offload_issue.md`](docs/ruler_32k_chunked_offload_issue.md) | ⚠️ OPEN ISSUE: 32K chunked offload accuracy problem (20% error rate in RULER) |
| [`docs/chunked_attention_solutions.md`](docs/chunked_attention_solutions.md) | 🔧 SOLUTIONS: Chunked attention 准确性问题的代码分析和解决方案 |
## GPU Mutex for Multi-Instance Debugging