Commits

Branch selector

main

User selector

All users

All time

Commit history

Commits on Jun 20, 2026

• [codex] Preserve reasoning-only Responses output (#1774)

  Show description for 32ae73b
  xeophon

  authored
  32ae73bView commit details

  Copy full SHA for 32ae73b

  Browse repository at this point

Commits on Jun 19, 2026

• Allow reasoning-only model responses (#1767)

  Show description for 8f86e28
  xeophon

  authored
  8f86e28View commit details

  Copy full SHA for 8f86e28

  Browse repository at this point

Commits on Jun 15, 2026

• Add CLI-Gym Harbor taskset (#1665)

  Show description for 5ed2809
  rasdani

  authored
  5ed2809View commit details

  Copy full SHA for 5ed2809

  Browse repository at this point

Commits on Jun 12, 2026

• Add SandboxDebugEnv for sandbox tasksets (#1652)

  rasdani

  authored
  7a60937View commit details

  Copy full SHA for 7a60937

  Browse repository at this point

• Merge pull request #1561 from PrimeIntellect-ai/feat/harness-transform-prompt-hook

  Show description for fd28c14
  hubert-marek

  authored
  fd28c14View commit details

  Copy full SHA for fd28c14

  Browse repository at this point

• build: add fallback-version for no-VCS (Docker) editable builds (#1650)

  Show description for bf43a31
  

  JannikSt

  and

  eligotts

  authored
  bf43a31View commit details

  Copy full SHA for bf43a31

  Browse repository at this point

Commits on Jun 11, 2026

• Add Terminal-Lego composable taskset (#1582)

  Show description for f3c8281
  rasdani

  authored
  f3c8281View commit details

  Copy full SHA for f3c8281

  Browse repository at this point

• fix(v1): isolate sandbox client failures (#1622)

  xeophon

  authored
  d13465cView commit details

  Copy full SHA for d13465c

  Browse repository at this point

• fix(v1): cast awaited prepare_prompt result to satisfy ty

  Show description for 5b411a1
  

  hubert-marek

  and

  claude

  committed
  5b411a1View commit details

  Copy full SHA for 5b411a1

  Browse repository at this point

Commits on Jun 9, 2026

• feat: add QUEST open-ended scoring (#1580)

  Show description for ffcf6ce
  rasdani

  authored
  ffcf6ceView commit details

  Copy full SHA for ffcf6ce

  Browse repository at this point

• Add v1 tunnel labels (#1584)

  Show description for 5bf80a9
  parkerpettit

  authored
  5bf80a9View commit details

  Copy full SHA for 5bf80a9

  Browse repository at this point

• Default OpenSeeker full dataset and use filter_fn row filters (#1579)

  Show description for 0796112
  rasdani

  authored
  0796112View commit details

  Copy full SHA for 0796112

  Browse repository at this point

• Match REDSearcher released judge prompt (#1581)

  rasdani

  authored
  06efedfView commit details

  Copy full SHA for 06efedf

  Browse repository at this point

• fix(v1): extend sandbox readiness wait (#1577)

  xeophon

  authored
  c0c1f15View commit details

  Copy full SHA for c0c1f15

  Browse repository at this point

• fix(eval): keep saved results append-only (#1564)

  xeophon

  authored
  7e2c192View commit details

  Copy full SHA for 7e2c192

  Browse repository at this point

• Add REDSearcher search taskset (#1573)

  Show description for c013b1f
  rasdani

  authored
  c013b1fView commit details

  Copy full SHA for c013b1f

  Browse repository at this point

Commits on Jun 8, 2026

• add OpenSeeker search taskset (#1571)

  Show description for 0ad8b4d
  rasdani

  authored
  0ad8b4dView commit details

  Copy full SHA for 0ad8b4d

  Browse repository at this point

• Add QUEST search taskset (#1555)

  Show description for ffb5e01
  rasdani

  authored
  ffb5e01View commit details

  Copy full SHA for ffb5e01

  Browse repository at this point

• docs(v1): document Harness.prepare_prompt hook in BYO Harness guide

  Show description for dadff8e
  

  cursoragent

  and

  willccbb

  committed
  dadff8eView commit details

  Copy full SHA for dadff8e

  Browse repository at this point

• feat(v1): rename Harness.transform_prompt to prepare_prompt and apply in Runtime.submit_model_request

  Show description for 95efb23
  

  cursoragent

  and

  willccbb

  committed
  95efb23View commit details

  Copy full SHA for 95efb23

  Browse repository at this point

• feat(v1): add Harness.transform_prompt hook

  Show description for 4a99e8d
  

  hubert-marek

  and

  claude

  committed
  4a99e8dView commit details

  Copy full SHA for 4a99e8d

  Browse repository at this point

Commits on Jun 7, 2026

• feat(v1/sandbox): canonical /vf/model bridge (sandbox sends Messages; host resolves client + tokenizes) (#1560)

  Show description for a01ce52
  

  3 peopleauthored
  a01ce52View commit details

  Copy full SHA for a01ce52

  Browse repository at this point

Commits on Jun 6, 2026

• add dataset shuffle to `vf-eval` (#1556)

  rasdani

  authored
  98558a8View commit details

  Copy full SHA for 98558a8

  Browse repository at this point

• Tighten loose verifier runtime types (#1549)

  willccbb

  authored
  2477262View commit details

  Copy full SHA for 2477262

  Browse repository at this point

Commits on Jun 5, 2026

• Decouple composable harnesses from harness packages (#1551)

  xeophon

  authored
  6529849View commit details

  Copy full SHA for 6529849

  Browse repository at this point

• Merge harness version config and bump harnesses (#1544)

  xeophon

  authored
  7aa421bView commit details

  Copy full SHA for 7aa421b

  Browse repository at this point

• Use full checkout for uv CI jobs (#1553)

  xeophon

  authored
  dabc232View commit details

  Copy full SHA for dabc232

  Browse repository at this point

• Document SWE taskset progress (#1530)

  rasdani

  authored
  12d5ea7View commit details

  Copy full SHA for 12d5ea7

  Browse repository at this point

• Revert top-level v1 env config shorthand (#1548)

  willccbb

  authored
  a82248aView commit details

  Copy full SHA for a82248a

  Browse repository at this point

• RLM: remove legacy rlm (#1515)

  Show description for 069dcf8
  

  snimu

  and

  claude

  authored
  069dcf8View commit details

  Copy full SHA for 069dcf8

  Browse repository at this point

• [codex] Add replay taskset and harness (#1536)

  Show description for 50581e2
  

  willccbb

  and

  tim0120

  authored
  50581e2View commit details

  Copy full SHA for 50581e2

  Browse repository at this point

• Make v1 max_turns a stop signal with request-boundary admission (#1547)

  willccbb

  authored
  4cddaf5View commit details

  Copy full SHA for 4cddaf5

  Browse repository at this point

Commits on Jun 4, 2026

• Cache Hugging Face tokenizers in CI (#1542)

  xeophon

  authored
  3420677View commit details

  Copy full SHA for 3420677

  Browse repository at this point

• Add versioned agent install specs (#1506)

  xeophon

  authored
  f81b6f2View commit details

  Copy full SHA for f81b6f2

  Browse repository at this point

• [codex] Reject reasoning-only model responses (#1537)

  willccbb

  authored
  f54f189View commit details

  Copy full SHA for f54f189

  Browse repository at this point