[microsoft]ToRA is a series of Tool-integrated Reasoning LLM Agents designed to solve challenging mathematical reasoning problems by interacting with tools.
active 2023-10-08 → 2026-05-30 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 966 days from 2023-10-08 to 2026-05-30. Pushes: 23 total, peak 10 in a day. Pull requests: 2 total, peak 1 in a day. Issues: 31 total, peak 4 in a day. Comments: 13 total, peak 5 in a day. Stars: 638 total, peak 81 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 |
|---|---|---|---|---|
| ZubinGou | 42 | 21 | 1 | 10 |
| runzeer | 3 | 0 | 0 | 0 |
| AceCHQ | 2 | 0 | 0 | 0 |
| LiuTaowen-Tony | 2 | 0 | 0 | 1 |
| ZhihongShao | 2 | 2 | 0 | 0 |
| MatCaviar | 2 | 0 | 0 | 0 |
| xufangzhi | 1 | 0 | 0 | 0 |
| shuoyinn | 1 | 0 | 0 | 1 |
| Rakesh123Art | 1 | 0 | 0 | 0 |
| tongyx361 | 1 | 0 | 0 | 0 |
| dependabot[bot] | 1 | 0 | 1 | 0 |
| HumanGamer | 1 | 0 | 0 | 0 |
| Mucalinda2436 | 1 | 0 | 0 | 0 |
| cweiqiang | 1 | 0 | 0 | 0 |
| JianxinMa | 1 | 0 | 0 | 0 |
| yuchenlin | 1 | 0 | 0 | 1 |
| emrgnt-cmplxty | 1 | 0 | 0 | 0 |
| X-Lai | 1 | 0 | 0 | 0 |
| zhuxunyu | 1 | 0 | 0 | 0 |
| Pobby321 | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#32ahmedmustahid2025-07-22 09:48Would love to have an updated Tora model
- Issue#31Rakesh123Art2025-03-02 17:09Tora eepo
- Issue comment#27yuchenlin2024-09-15 06:56when the data will be open-sourced?
- Issue#29tongyx3612024-08-04 05:02What are the temperatures used for nucleus sampling in data synthesis?
- Issue#28AceCHQ2024-07-27 15:31Request for ToRA Self-Consistency Results with k=10 and k=20
- Issue#26Mucalinda24362024-06-16 12:21请问有propmt GPT按照TORA流程解决问题的代码吗？
- Issue#25JianxinMa2024-04-19 02:36Is ToRA-69k, the SFT dataset, open-sourced?
- Issue#24X-Lai2024-03-21 07:40How to generate ToRA Corpus based on scripts/infer_api.sh ?
- Issue comment#21LiuTaowen-Tony2024-01-21 10:25Clarification and Issues Reproducing ToRA-Code-34b Results
- Issue comment#20ZubinGou2024-01-21 06:32Could you please provide the results based on more powerful base model llemma_7b?
- Issue#20ZubinGou2024-01-21 06:32Could you please provide the results based on more powerful base model llemma_7b?
- Issue comment#21ZubinGou2024-01-21 06:29Clarification and Issues Reproducing ToRA-Code-34b Results
- Issue#21LiuTaowen-Tony2024-01-20 20:26Reconfirm Deepspeed setting and CodeLlama base model used by ToRA
- Issue#16ZubinGou2024-01-03 06:48Fail to reproduce the codellama baseline with the code base
- Issue comment#18shuoyinn2024-01-03 05:28About the Imitation Learning in the paper
- Issue comment#19ZubinGou2024-01-03 04:32Hyperparameter of Self-consistency?
- Issue#19ZubinGou2024-01-03 04:32Hyperparameter of Self-consistency?
- Issue comment#18ZubinGou2024-01-03 04:26About the Imitation Learning in the paper
- Issue#18ZubinGou2024-01-03 04:26About the Imitation Learning in the paper
- Issue comment#16ZubinGou2024-01-03 04:20Fail to reproduce the codellama baseline with the code base
- Issue comment#15ZubinGou2024-01-03 04:19long context input
- Issue#15ZubinGou2024-01-03 04:19long context input
- Issue#19choco99662024-01-02 13:22Hyperparameter of Self-consistency?
- Pull request#17dependabot[bot]2023-12-20 21:11
- Issue#14ZubinGou2023-12-17 11:28Issues installing flash_attention==2.0.1, stuck & also freezes screen on local computer
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 638 stars here means stars gained during the window, not the repo's star count.