Benchmark Almanac

Snorkel Agentic Coding

Claude Opus 4.6 65.2% leads since February 2026
Not in model cards

Agents complete 100 multi-step coding tasks in sandboxed containers, spread evenly over four difficulty tiers. Unit tests and rubrics grade both outputs and the agent trajectory.

Scored as Pass@5 on the Harbor harness with a 30-minute cap per task. Snorkel archived the board after Agentic Coding 2.0.

Made by
Snorkel AI
Released
January 2026
Kind
Building whole projects
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Strongest model tested
Claude Opus 4.6 #37 on the capability index
Last model added
February 2026
Scores refreshed
November 12th, 2026
11
Models scored on its own leaderboard

Claude Opus 4.6 leads with 65.2% among models released by February 2026

Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in February 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 65.2 Claude Opus 4.6
  2. 58.0 Claude Opus 4.5
  3. 57.6 Claude Sonnet 4.5
  4. 51.6 Gemini 3 Pro
  5. 49.4 GPT-5.2
  6. 45.2 GPT-5
  7. 36.8 Kimi K2 Thinking
  8. 33.2 Devstral 2
  9. 25.2 Grok 4.1 Fast
  10. 18.8 Qwen3-Coder-480B-A35B
  11. 13.8 Mistral Large 3
  1. 1 Claude Opus 4.6 Pass@5 65.2%
  2. 2 Claude Opus 4.5 Pass@5 58%
  3. 3 Claude Sonnet 4.5 Pass@5 57.6%
  4. 4 Gemini 3 Pro Pass@5 51.6%
  5. 5 GPT-5.2 Pass@5 49.4%
  6. 6 GPT-5 Pass@5 45.2%
  7. 7 Kimi K2 Thinking Pass@5 36.8%
  8. 8 Devstral 2 Pass@5 33.2%
  9. 9 Grok 4.1 Fast Pass@5 25.2%
  10. 10 Qwen3-Coder-480B-A35B Pass@5 18.8%
  11. 11 Mistral Large 3 Pass@5 13.8%

Source: Official Snorkel Agentic Coding leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 4.6 65.2% Pass@5 Feb 2026 Official leaderboard
2 Claude Opus 4.5 58% Pass@5 Nov 2025 Official leaderboard
3 Claude Sonnet 4.5 57.6% Pass@5 Sep 2025 Official leaderboard
4 Gemini 3 Pro 51.6% Pass@5 Nov 2025 Official leaderboard
5 GPT-5.2 49.4% Pass@5 Dec 2025 Official leaderboard
6 GPT-5 45.2% Pass@5 Aug 2025 Official leaderboard
7 Kimi K2 Thinking 36.8% Pass@5 Nov 2025 Official leaderboard
8 Devstral 2 33.2% Pass@5 Dec 2025 Official leaderboard
9 Grok 4.1 Fast 25.2% Pass@5 Nov 2025 Official leaderboard
10 Qwen3-Coder-480B-A35B 18.8% Pass@5 Jul 2025 Official leaderboard
11 Mistral Large 3 13.8% Pass@5 Dec 2025 Official leaderboard

2 models have held number one on the official leaderboard since its release in January 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Own leaderboard Not tested since Feb 2026
Claude Opus 4.6 65.2%

Source: Official Snorkel Agentic Coding leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.6 Claude Opus 4.5 Claude Sonnet 4.5 GPT-5 Qwen3-Coder-480B-A35B 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.6 Claude Opus 4.5

Source: Official Snorkel Agentic Coding leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

Anthropic OpenAI
Snorkel Agentic Coding 2.0 3 leaders
GPT-6 Astra 47.6%
Snorkel Agentic Coding Not tested since Feb 2026
Claude Opus 4.6 65.2%
Snorkel Agentic Coding 2.0 1 leader
GPT-6 Astra 47.6%

Source: Official leaderboards

Replaced by Snorkel Agentic Coding 2.0

More in Coding

All 282 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471