Benchmark Almanac

Vals RSI Index

Claude Fable 5.2 0.38 leads since November 2026
Not in model cards

Agents work four open AI R&D problems under fixed compute and time budgets: compression, LM training, judge harness work and post-training. Each runs in its vendor harness.

Each task maps a measured result, such as bits per byte on enwik8, onto a score between a baseline and a theoretical best. 25 models on October 15th, 2026.

Made by
Vals AI
Released
August 2026
Kind
AI research tasks
Level
Research frontier
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Strongest model tested
Claude Fable 5.2 #1 on the capability index
Last model added
November 2026
Scores refreshed
November 12th, 2026
27
Models scored on its own leaderboard

Claude Fable 5.2 leads with 0.38

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. Every run is in the next chart.

27 models. Scroll sideways for the rest.

  1. 0.38 Claude Fable 5.2
  2. 0.37 Claude Opus 5.5
  3. 0.36 Claude Fable 5.1
  4. 0.33 Claude Opus 5
  5. 0.30 Gemini 4 Argon
  6. 0.28 GPT-6 Sol
  7. 0.28 Gemini 4 Carbon
  8. 0.27 GPT-6 Astra
  9. 0.27 GPT-6.1 Sol
  10. 0.26 Grok 4.6
  11. 0.24 Muse Wonder Full
  12. 0.24 Claude Fable 5
  13. 0.24 Claude Opus 4.8
  14. 0.24 GPT-5.6 Sol
  15. 0.23 GLM-5.3
  16. 0.22 Qwen 3.8 Max
  17. 0.21 Kimi K3
  18. 0.20 Grok 4.7
  19. 0.20 Gemini 3.8 Flash
  20. 0.20 Muse Spark 1.3
  21. 0.19 Claude Opus 4.7
  22. 0.18 Gemini 3.7 Flash
  23. 0.17 GPT-5.5
  24. 0.16 GPT-5.2
  25. 0.15 GPT-5.4
  26. 0.15 Claude Opus 4.5
  27. 0.07 Kimi K2.5
  1. 1 Claude Fable 5.2 max 0.38
  2. 2 Claude Opus 5.5 max 0.37
  3. 3 Claude Fable 5.1 max 0.36
  4. 4 Claude Opus 5 max 0.33
  5. 5 Gemini 4 Argon high 0.30
  6. 6 GPT-6 Sol max 0.28
  7. 7 Gemini 4 Carbon high 0.28
  8. 8 GPT-6 Astra max 0.27
  9. 9 GPT-6.1 Sol max 0.27
  10. 10 Grok 4.6 xhigh 0.26
  11. 11 Muse Wonder Full max 0.24
  12. 12 Claude Fable 5 max 0.24
  13. 13 Claude Opus 4.8 max 0.24
  14. 14 GPT-5.6 Sol max 0.24
  15. 15 GLM-5.3 max 0.23
  16. 16 Qwen 3.8 Max max 0.22
  17. 17 Kimi K3 max 0.21
  18. 18 Grok 4.7 xhigh 0.20
  19. 19 Gemini 3.8 Flash high 0.20
  20. 20 Muse Spark 1.3 max 0.20
  21. 21 Claude Opus 4.7 max 0.19
  22. 22 Gemini 3.7 Flash high 0.18
  23. 23 GPT-5.5 xhigh 0.17
  24. 24 GPT-5.2 xhigh 0.16
  25. 25 GPT-5.4 xhigh 0.15
  26. 26 Claude Opus 4.5 non-thinking 0.15
  27. 27 Kimi K2.5 max 0.07

Source: Official Vals RSI Index leaderboard

#ModelScoreSettingDateSource
1 Claude Fable 5.2 0.38 max, Claude Code Nov 2026 Official leaderboard
2 Claude Opus 5.5 0.37 max, Claude Code Sep 2026 Official leaderboard
3 Claude Fable 5.1 0.36 max, Claude Code Sep 2026 Official leaderboard
4 Claude Opus 5 0.33 max, Claude Code Jul 2026 Official leaderboard
5 Gemini 4 Argon 0.30 high, Antigravity Sep 2026 Official leaderboard
6 GPT-6 Sol 0.28 max, Codex Sep 2026 Official leaderboard
7 Gemini 4 Carbon 0.28 high, Antigravity Oct 2026 Official leaderboard
8 GPT-6 Astra 0.27 max, Codex Sep 2026 Official leaderboard
9 GPT-6.1 Sol 0.27 max, Codex Sep 2026 Official leaderboard
10 Grok 4.6 0.26 xhigh, Grok Build Aug 2026 Official leaderboard
11 Muse Wonder Full 0.24 max, Muse Code Oct 2026 Official leaderboard
12 Claude Fable 5 0.24 max, Claude Code Jun 2026 Official leaderboard
13 Claude Opus 4.8 0.24 max, Claude Code May 2026 Official leaderboard
14 GPT-5.6 Sol 0.24 max, Codex Jul 2026 Official leaderboard
15 GLM-5.3 0.23 max, Claude Code Aug 2026 Official leaderboard
16 Qwen 3.8 Max 0.22 max, Qwen Code Aug 2026 Official leaderboard
17 Kimi K3 0.21 max, Kimi Code Jul 2026 Official leaderboard
18 Grok 4.7 0.20 xhigh, Grok Build Sep 2026 Official leaderboard
19 Gemini 3.8 Flash 0.20 high, Antigravity Sep 2026 Official leaderboard
20 Muse Spark 1.3 0.20 max, Muse Code Sep 2026 Official leaderboard
21 Claude Opus 4.7 0.19 max, Claude Code Apr 2026 Official leaderboard
22 Gemini 3.7 Flash 0.18 high, Antigravity Aug 2026 Official leaderboard
23 GPT-5.5 0.17 xhigh, Codex Apr 2026 Official leaderboard
24 GPT-5.2 0.16 xhigh, Codex Dec 2025 Official leaderboard
25 GPT-5.4 0.15 xhigh, Codex Mar 2026 Official leaderboard
26 Claude Opus 4.5 0.15 non-thinking, Claude Code Nov 2025 Official leaderboard
27 Kimi K2.5 0.07 max, Kimi Code Jan 2026 Official leaderboard

The best score, 0.38, costs $384 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

Moonshot AI xAI Z.ai Anthropic Meta OpenAI Google DeepMind Cost frontier One model at each effort level
0.00 0.10 0.20 0.30 0.40 $10 $100 $1,000 $10,000 Cost per task, log scale Claude Fable 5.2, max Claude Fable 5.1, max Gemini 4 Argon, high GPT-6.1 Sol, max GLM-5.3, max Muse Spark 1.3, max Kimi K2.5, max 0.00 0.10 0.20 0.30 0.40 $10 $100 $1,000 $10,000 Cost per task, log scale Claude Fable 5.2, max Gemini 4 Argon, high

Source: Official Vals RSI Index leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Fable 5.2 max Claude Code 0.38 $384
Claude Opus 5.5 max Claude Code 0.37 $412
Claude Fable 5.1 max Claude Code 0.36 $298
Claude Opus 5 max Claude Code 0.33 $419
Gemini 4 Argon high Antigravity 0.30 $180
GPT-6 Sol max Codex 0.28 $665
Gemini 4 Carbon high Antigravity 0.28 $231
GPT-6 Astra max Codex 0.27 $760
GPT-6.1 Sol max Codex 0.27 $98
Grok 4.6 xhigh Grok Build 0.26 $88
Muse Wonder Full max Muse Code 0.24
Claude Fable 5 max Claude Code 0.24 $491
Claude Opus 4.8 max Claude Code 0.24 $465
GPT-5.6 Sol max Codex 0.24 $347
GLM-5.3 max Claude Code 0.23 $62
Qwen 3.8 Max max Qwen Code 0.22 $114
Kimi K3 max Kimi Code 0.21 $137
Grok 4.7 xhigh Grok Build 0.20 $237
Gemini 3.8 Flash high Antigravity 0.20 $332
Muse Spark 1.3 max Muse Code 0.20 $49
Claude Opus 4.7 max Claude Code 0.19 $1,003
Gemini 3.7 Flash high Antigravity 0.18 $180
GPT-5.5 xhigh Codex 0.17 $857
GPT-5.2 xhigh Codex 0.16 $102
GPT-5.4 xhigh Codex 0.15 $287
Claude Opus 4.5 non-thinking Claude Code 0.15 $366
Kimi K2.5 max Kimi Code 0.07 $25

4 models have held number one on the official leaderboard since its release in August 2026

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

Anthropic
Own leaderboard 4 leaders
Claude Fable 5.2 0.38

Source: Official Vals RSI Index 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
0.00 0.10 0.20 0.30 0.40 2026 2027 0.00 0.10 0.20 0.30 0.40 2026 2027

Source: Official Vals RSI Index 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