benchlm.ai frontier composite
Third-party composite frontier ranking
composite indexactive
Score, benchlm.ai frontier top 6
Definition
- Version
- not in the notes
- Category
- composite index
- Maintainer
- benchlm.ai
- Status
- active
- Home page
- https://benchlm.ai/frontier-ai-models
Sources
- Frontier AI Models: Live Top 10 Rankings, Evidence and Pricing (October 2026) benchlm.ai https://benchlm.ai/frontier-ai-models
Scores at a Glance
Score, benchlm.ai frontier top 6
Higher is better for this unit.
1
Claude Opus 5.5Anthropic
GPT-6 AstraOpenAIClaude Sonnet 5.5AnthropicClaude Fable 5.1Anthropic
Gemini 4 ArgonGoogle DeepMindestimatedGPT-6.1 SolOpenAIestimated
Claude Opus 5.5Anthropic86.4 index
vendor-reported2GPT-6 AstraOpenAI85 index
vendor-reported3Claude Sonnet 5.5Anthropic83.9 index
vendor-reported4Claude Fable 5.1Anthropic82 index
vendor-reported5Gemini 4 ArgonGoogle DeepMindestimated81.8 index
vendor-reported6GPT-6.1 SolOpenAIestimated81.5 index
vendor-reported0255075100
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| Claude Opus 5.5 | 86.4 index | vendor-reported | – | benchlm.ai frontier top 6 | 9 Oct 2026 | benchlm.ai |
| GPT-6 Astra | 85 index | vendor-reported | – | benchlm.ai frontier top 6 | 9 Oct 2026 | benchlm.ai |
| Claude Sonnet 5.5 | 83.9 index | vendor-reported | – | benchlm.ai frontier top 6 | 9 Oct 2026 | benchlm.ai |
| Claude Fable 5.1 | 82 index | vendor-reported | – | benchlm.ai frontier top 6 | 9 Oct 2026 | benchlm.ai |
| Gemini 4 Argon | 81.8 index | vendor-reported | estimated | benchlm.ai frontier top 6 | 9 Oct 2026 | benchlm.ai |
| GPT-6.1 Sol | 81.5 index | vendor-reported | estimated | benchlm.ai frontier top 6 | 9 Oct 2026 | benchlm.ai |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "benchlm.ai Frontier Composite", https://aurion.semai.id/benchmarks/benchlm-composite/, records of October 2026.