Skip to content

Model router for agentic systems. Routes every prompt to the right model in <50ms. Cut costs 40-70% with just an endpoint change.

active 2026-05-152026-09-02 (UTC)

Complete coverage26,770 / 26,770 hourly files (100%) · 2 absent upstream2023-08-152026-09-03 (UTC)
Events
1.2K
Pushes
549
Pull requests
76
Issues
1
Stars
53
Forks
3

Activity over time

Daily event counts in the loaded window

Line chart, 111 days from 2026-05-15 to 2026-09-02. Pushes: 549 total, peak 44 in a day. Pull requests: 76 total, peak 18 in a day. Issues: 1 total, peak 1 in a day. Comments: 85 total, peak 12 in a day. Stars: 53 total, peak 13 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

Recent activity

Latest issues, pull requests and releases

  • Pull request#1135a-churchill2026-09-01 05:19
  • Pull request#1096a-churchill2026-08-28 06:41
  • Pull request#1069a-churchill2026-08-27 03:39
  • Pull request#1058devin-ai-integration[bot]2026-08-26 21:41
  • Issue comment#1028claude[bot]2026-08-25 02:49
    feat(proxy): escalate struggling sessions up a cluster before sideways
  • Pull request#981steventohme2026-08-21 17:30
  • Pull request#948aminsamir452026-08-18 23:43
  • Issue comment#946devin-ai-integration[bot]2026-08-18 23:01
    Degrade to a same-cluster model when the routed model's bindings are exhausted
  • Pull request#939steventohme2026-08-18 21:13
  • Issue comment#938claude[bot]2026-08-18 21:04
    fix: price cache planner by provider binding
  • Issue comment#910a-churchill2026-08-14 03:24
    Show routing badge only when the served model changes
  • Issue comment#900devin-ai-integration[bot]2026-08-12 13:11
    feat(providers): remap outbound model IDs for deployment-key endpoints
  • Issue comment#885claude[bot]2026-08-07 21:15
    feat(auth): add analytics_read key scope with ra_ prefix
  • Issue comment#834claude[bot]2026-07-24 19:12
    fix(translate): keep CC scheduling tools on cross-vendor emit
  • Issue comment#833claude[bot]2026-07-24 18:41
    feat(catalog): add claude-opus-5, retire opus-4-8 from routing (v0.75)
  • Issue comment#743claude[bot]2026-07-16 00:24
    fix: prioritize explicit force-model pins
  • Issue comment#168devin-ai-integration[bot]2026-07-15 23:02
    docs: add chinese README.zh-CN from README
  • Issue comment#738devin-ai-integration[bot]2026-07-15 23:00
    fix(providers/google): buffer NativeClient error responses for failover
  • Issue#719rohith5002026-07-14 18:39
    HMM router-feedback (/feedback path) training delta is byte-unbounded same bug class as #715/#716, different code path
  • Issue comment#699claude[bot]2026-07-10 21:30
    Send HMM feedback training transcript
  • Issue comment#689claude[bot]2026-07-10 18:59
    feat: persist policy router rollout controls
  • Issue comment#692greptile-apps[bot]2026-07-10 04:25
    feat: expose policy candidate provider bindings
  • Issue comment#613claude[bot]2026-07-03 20:51
    chore: guard against provider-dispatch drift with test + onboarding doc
  • Pull request#585steventohme2026-07-03 16:29
  • Issue comment#572greptile-apps[bot]2026-07-02 15:03
    fix(proxy): clear suppressed Claude subscription so failover uses the deployment key

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 53 stars here means stars gained during the window, not the repo's star count.