Benchmark Almanac

KernelQuest

Claude Opus 5.5 79% leads since September 2026
Not in model cards

Agents optimize Triton kernels. A task passes when the kernel produces correct outputs and reaches the speed target on the same GPU.

Engineers demonstrated the target speed for each of the 100 tasks.

Made by
Turing
Released
September 2026
Kind
GPU kernels
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Measures
Capability
In and out
Text, Code → Code, Actions
Interaction
Agentic
Environment
Code sandbox
Refresh
One fixed set
Run by
A host
In views
Agents
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
4
Models scored on its own leaderboard

Claude Opus 5.5 leads with 79% among models released by September 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

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

  1. 1 Claude Opus 5.5 79%
  2. 2 GPT-6 Astra 43.3%
  3. 3 Gemini 3.8 Flash 14.7%
  4. 4 Grok 4.7 11.3%

Source: Official KernelQuest leaderboard

#ModelScore ±SettingDateSource
1 Claude Opus 5.5 79% ±4.6 Claude Code Sep 2026 Official leaderboard
2 GPT-6 Astra 43.3% ±5.6 Codex Sep 2026 Official leaderboard
3 Gemini 3.8 Flash 14.7% ±4 Antigravity CLI Sep 2026 Official leaderboard
4 Grok 4.7 11.3% ±3.6 Grok Build Sep 2026 Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as L1: Single operator, L2: Fused epilogue and L3: Complete model. Claude Opus 5.5 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

L1: Single operator

4 models we track have a score here.

  1. 1 Claude Opus 5.5 75%
  2. 2 GPT-6 Astra 52%
  3. 3 Grok 4.7 ↑1 17%
  4. 4 Gemini 3.8 Flash ↓1 12%

3 models have held number one on the official leaderboard since its release in September 2026

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

Google DeepMind OpenAI Anthropic
Own leaderboard Not tested since Sep 2026

Source: Official KernelQuest leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% Near the ceiling (85%) Claude Opus 5.5 GPT-6 Astra Gemini 3.8 Flash 0% 20% 40% 60% 80% 100% Near the ceiling (85%) Claude Opus 5.5 GPT-6 Astra

Source: Official KernelQuest leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in AI research

All 164 in AI research
  • PaperBench In use Agent replicates ICML 2024 papers from scratch. LLM judge grades against author-written hierarchical rubrics. Qwen 3.8 Max 93%
  • MLS-Bench In use Agent must invent one algorithmic change in an ML research scaffold whose gain transfers across settings. Scored relative to human baselines. Claude Fable 5.1 50.3%
  • PostTrainBench 1.2 In use CLI agent gets a base model, one H100 and a time budget to post-train it for a target benchmark. Scored on normalized gain. Claude Fable 5.1 44.6%
  • MLE-bench In use Kaggle-style ML engineering competitions. Agent trains models and is scored against competition leaderboards. GPT-6.2 Astra 96.1%
  • PMPP-Hard In use Agents write GPU kernels for 69 tasks that must be correct, and on 48 of them also run within 1.25 times the reference time. The score is the area under the solved-tasks curve over a 10-hour budget, out of 690 task-hours. Kimi K3 0.71
  • KernelBench In use Replace PyTorch operators with faster custom GPU kernels. Scored on correctness and speedup. MiniMax-M3 28.8%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (4 entries).
  • Note Models scored Only 4 models we track.
  • Pass Newest score The newest is from September 2026, for Claude Opus 5.5.
  • Note Frontier models Only 2 of the top 10 have a score. The strongest tested is GPT-6 Astra (#4).
  • Pass Headroom The best is 79%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on labs.turing.com: 4 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check