The code of our paper "InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory"(清华超长上下文记忆)
active 2024-03-03 → 2026-05-09 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 798 days from 2024-03-03 to 2026-05-09. Pushes: 12 total, peak 2 in a day. Pull requests: 11 total, peak 3 in a day. Issues: 88 total, peak 7 in a day. Comments: 120 total, peak 16 in a day. Stars: 366 total, peak 74 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| guyan364 | 78 | 12 | 7 | 43 |
| ChuanhongLi | 23 | 0 | 0 | 12 |
| Minami-su | 21 | 0 | 1 | 14 |
| xcjthu | 9 | 0 | 0 | 5 |
| Michelleable | 9 | 0 | 0 | 7 |
| huliangbing | 8 | 0 | 0 | 6 |
| umie0128 | 6 | 0 | 0 | 2 |
| Becomebright | 5 | 0 | 0 | 3 |
| thistleknot | 4 | 0 | 0 | 2 |
| yuanyehome | 4 | 0 | 0 | 3 |
| cat-sun | 4 | 0 | 0 | 0 |
| ehuaa | 4 | 0 | 0 | 2 |
| NaivePawn | 4 | 0 | 0 | 3 |
| yinochaos | 4 | 0 | 0 | 3 |
| MrJiangZhongZheng | 3 | 0 | 0 | 1 |
| tiaotiaosong | 3 | 0 | 0 | 1 |
| KangcongLi | 3 | 0 | 0 | 1 |
| kristaller486 | 3 | 0 | 0 | 1 |
| DefinitlyEvil | 3 | 0 | 0 | 2 |
| fst813 | 3 | 0 | 0 | 2 |
Recent activity
Latest issues, pull requests and releases
- Pull request#63AshleyEmily2026-03-15 08:36
- Pull request#63AshleyEmily2026-03-15 08:35
- Pull request#62psi2412025-12-10 00:13
- Issue#59umie01282025-05-30 07:06流式输出显存爆炸
- Issue#59umie01282025-05-20 04:00流式输出显存爆炸
- Issue#59umie01282025-05-18 13:56流式输出显存爆炸
- Issue#59umie01282025-05-16 06:47流式输出显存爆炸
- Issue#58qaqjx2025-03-03 07:12'MistralAttention' object has no attribute 'rotary_emb'
- Issue#57KangcongLi2025-02-22 07:44problem when reproducing Mistral-based InfLLM
- Issue comment#57KangcongLi2025-02-22 07:44problem when reproducing Mistral-based InfLLM
- Issue comment#57umie01282025-02-12 08:01problem when reproducing Mistral-based InfLLM
- Issue#57KangcongLi2025-01-20 08:20problem when reproducing Mistral-based InfLLM
- Issue comment#41umie01282025-01-17 08:47可以与vllm集成吗？
- Issue#56sherlcok3141592025-01-06 03:28[Question] 运行速度很慢
- Issue comment#51DefinitlyEvil2024-11-12 12:34Llama 3.1 Load Fail
- Issue comment#51yuanyehome2024-11-12 06:16Llama 3.1 Load Fail
- Issue comment#51DefinitlyEvil2024-11-11 16:12Llama 3.1 Load Fail
- Issue comment#51yuanyehome2024-11-11 03:02Llama 3.1 Load Fail
- Issue comment#55yuanyehome2024-11-08 03:28Error when reproducing mistral results
- Issue comment#55guyan3642024-11-08 03:24Error when reproducing mistral results
- Issue#55yuanyehome2024-11-08 03:10Error when reproducing mistral results
- Issue#50Lu-kuan-lpk2024-10-31 10:35Implementation of Streaming-llm
- Issue#54FranciscoPark2024-10-30 06:51Block-Level Memory Units representative tokens
- Issue#54FranciscoPark2024-10-28 08:54Block-Level Memory Units representative tokens
- Issue#53chaiyixuan2024-10-26 07:10请问该算法可以支持训练么？
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 366 stars here means stars gained during the window, not the repo's star count.