LM Market Cap LMC Score: leaderboard
A composite index from lmmarketcap.com, not a task models run. Published results on benchmarks such as Arena Elo, MMLU, GPQA, HumanEval, SWE-bench and MATH are rescaled to a common range and averaged for ninety percent of the score, with feature breadth and context window size as tiebreakers. It is recomputed hourly and runs from 0 to 100, higher is better; variants such as batch endpoints appear as separate rows.
Metric: LMC Score (0-100). Source: lmmarketcap.com. Saturation forecast: Around February 2027. 217 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 95.9 | #20 |
| 2 | Claude Fable 5.1 | 95.9 | #10 |
| 3 | Claude Opus 4.7 | 95.1 | #51 |
| 4 | Claude Opus 4.8 | 95.1 | #39 |
| 5 | Claude Opus 5 | 95.1 | #16 |
| 6 | Claude Opus 5.5 | 95.1 | #3 |
| 7 | GPT-5.5 | 92.7 | #33 |
| 8 | GPT-5.5 Pro | 92.7 | #23 |
| 9 | Gemini 3.1 Pro (Preview) | 92.2 | #64 |
| 10 | GPT-5.4 | 91.9 | #83 |
| 11 | GPT-5.4 Pro (xHigh) | 91.9 | #38 (GPT-5.4 Pro) |
| 12 | GPT-5.2 | 90.5 | #115 |
| 13 | GPT-5.3 Codex | 90.5 | #72 |
| 14 | GPT-5.2 Instant | 90.5 | #217 |
| 15 | GPT-5.2 Pro | 90.5 | #70 |
No result here: #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=lm-market-cap-lmc-score · How It Works · Data refreshed daily, snapshot 2026-11-05.