Benchmark Almanac

OpenRSI Index Marin Scaling Ladder

GPT-5.6 Sol 1.01
Not in model cards

Agents design one optimizer for six model sizes. The score is the published complete six-rung reward relative to AdamH.

One track of OpenRSI Index. The OpenRSI Index record keeps lab scores that combine tracks. Published measurements include a result that failed staging guards.

Released
September 2026
Kind
AI research tasks
Level
Research frontier
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Strongest model tested
Claude Opus 5 #10 on the capability index
Last model added
July 2026
Scores refreshed
November 12th, 2026
2
Models scored on its own leaderboard

GPT-5.6 Sol leads with 1.01

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.

  1. 1.01 GPT-5.6 Sol
  2. 1.00 Claude Opus 5
  1. 1 GPT-5.6 Sol measured; staged 1.01
  2. 2 Claude Opus 5 measured; staging guards failed 1.00

Source: Official OpenRSI Index Marin Scaling Ladder leaderboard

#ModelScoreSettingDateSource
1 GPT-5.6 Sol 1.01 measured; staged, Codex Jul 2026 Official leaderboard
2 Claude Opus 5 1.00 measured; staging guards failed, Claude Code Jul 2026 Official 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
1.00 1.00 1.00 1.00 1.01 1.01 1.00 1.00 1.00 1.00 1.01 1.01

Source: Official OpenRSI Index Marin Scaling Ladder 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.

More in AI research

All 104 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 Astra 93.8%
  • KernelBench In use Replace PyTorch operators with faster custom GPU kernels. Scored on correctness and speedup. MiniMax-M3 28.8%
  • 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. Inkling 2 Small 494.3