LLM Router is an open-source project for building, training, and running a transformer-based router that picks the best LLM for each query. It helps balance cost and quality of the responses. The repo includes everything you need for training, evaluation, and running a router in chat applications.
active 2025-03-23 → 2025-10-20 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 212 days from 2025-03-23 to 2025-10-20. Pushes: 109 total, peak 14 in a day. Pull requests: 18 total, peak 4 in a day. Issues: 14 total, peak 3 in a day. Comments: 2 total, peak 1 in a day. Stars: 4 total, peak 1 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 |
|---|---|---|---|---|
| Amir-Mohseni | 117 | 83 | 16 | 2 |
| rej100 | 14 | 14 | 0 | 0 |
| cristic0j | 10 | 9 | 1 | 0 |
| franbvc | 1 | 1 | 0 | 0 |
| Vicente711 | 1 | 1 | 0 | 0 |
| dependabot[bot] | 1 | 0 | 1 | 0 |
| CARREY-HUANG | 1 | 1 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#21Amir-Mohseni2025-09-16 10:57
- Pull request#20Amir-Mohseni2025-09-16 10:51
- Pull request#19Amir-Mohseni2025-08-06 19:11
- Pull request#19Amir-Mohseni2025-08-06 19:11
- Pull request#18Amir-Mohseni2025-08-06 19:06
- Pull request#18Amir-Mohseni2025-08-06 19:06
- Issue#17Amir-Mohseni2025-06-08 11:23Add openai compatible server for router
- Issue#17Amir-Mohseni2025-05-29 09:06Add openai compatible server for router
- Pull request#15Amir-Mohseni2025-05-27 17:02
- Pull request#15cristic0j2025-05-12 23:55
- Issue#7Amir-Mohseni2025-05-12 16:44Setup RouteLLM for comparisons
- Issue#12Amir-Mohseni2025-05-10 10:16Qwen3-14B Data Extraction
- Issue#14Amir-Mohseni2025-05-10 10:16Extract Val/Test Splits (k = 3)
- Issue#13Amir-Mohseni2025-05-08 13:32Extract Train Split responses (k=3)
- Issue#13Amir-Mohseni2025-05-08 12:20Extract Train Split responses (k=3)
- Issue#6Amir-Mohseni2025-05-08 11:40Extract model responses from MMLU Pro and convert to our current formatting
- Issue#9Amir-Mohseni2025-05-03 12:51Add tensor parallel option for inferencing vllm on multiple GPUs
- Pull request#11Amir-Mohseni2025-05-03 12:51
- Pull request#11Amir-Mohseni2025-05-03 12:50
- Issue#10Amir-Mohseni2025-05-03 12:37Script for pushing model responses to huggingface hub
- Issue#9Amir-Mohseni2025-05-03 12:36Add tensor parallel option for inferencing vllm on multiple GPUs
- Issue#8Amir-Mohseni2025-04-30 14:08Setup Webapp Interface (Similar to GPT)
- Issue#7Amir-Mohseni2025-04-30 14:08Setup RouteLLM for comparisons
- Issue#6Amir-Mohseni2025-04-30 14:08Extract model responses from MMLU Pro and convert to our current formatting
- Pull request#5dependabot[bot]2025-04-29 15:54
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 4 stars here means stars gained during the window, not the repo's star count.