LLM Leaderboard
Live ranking of frontier language models by BenchLM composite score, with published API output pricing.
Source: BenchLM.ai · last verified November 19, 2026 · 214 ranked models tracked · top 30 shown here.
Data is CC BY-NC 4.0, © BenchLM.ai. This page is regenerated from their public leaderboard.
Score vs. cost
- Proprietary (25)
- Open Weight (4)
- Pending (1)
- Cost-efficient frontier
Top 30
| # | Model | Provider | Type | Context | Score | Output $/1M |
|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.2 | Anthropic | Proprietary | 1M | 93.2 | $40.00 |
| 2 | GPT-6.2 Astra | OpenAI | Proprietary | 1.05M | 91.1 | $40.00 |
| 3 | Gemini 4 Carbon | Proprietary | 1M | 88.1 | $10.00 | |
| 4 | Gemini 4 Neon | Proprietary | 1M | 86.3 | $15.00 | |
| 5 | GPT-6 Astra | OpenAI | Proprietary | 1.05M | 83.2 | $50.00 |
| 6 | Claude Opus 5.5 | Anthropic | Proprietary | 1M | 82.7 | $20.00 |
| 7 | GPT-6.2 Sol | OpenAI | Proprietary | 1.05M | 82.6 | $10.00 |
| 8 | GPT-6.1 Sol | OpenAI | Proprietary | 1.05M | 80.7 | $10.00 |
| 9 | GPT-6.1 Terra | OpenAI | Proprietary | 1.05M | 80.4 | $5.00 |
| 10 | Gemini 4 Argon | Proprietary | 1M | 79.1 | $10.00 | |
| 11 | Gemini 4 Hydrogen | Proprietary | 1M | 78.8 | $7.50 | |
| 12 | Claude Fable 5.1 | Anthropic | Proprietary | 1M | 78.7 | $50.00 |
| 13 | Claude Sonnet 5.5 | Anthropic | Proprietary | 1M | 77.9 | $10.00 |
| 14 | Muse Wonder Full | Meta | Proprietary | 1M | 77.2 | — |
| 15 | Claude Opus 5 | Anthropic | Proprietary | Not listed | 74.9 | $25.00 |
| 16 | Claude Fable 5 | Anthropic | Proprietary | 1M+ | 74.6 | $50.00 |
| 17 | Qwen4 Max | Alibaba | Proprietary | 1M | 74.5 | $4.50 |
| 18 | GPT-6 Sol | OpenAI | Proprietary | 1.05M | 73.3 | $10.00 |
| 19 | GPT-5.6 Sol | OpenAI | Proprietary | 1.05M | 73.0 | $20.00 |
| 20 | Muse Wonder | Meta | Proprietary | 1M | 71.2 | $20.00 |
| 21 | Kimi K3.1 Pro | Moonshot AI | Open Weight | 1.05M | 70.6 | $10.00 |
| 22 | GPT-5.5 Pro | OpenAI | Proprietary | 1M | 70.1 | $180.00 |
| 23 | DeepSeek V4.1 Flash 1017 | DeepSeek | Open Weight | 1M | 69.9 | $1.20 |
| 24 | GPT-5.5 | OpenAI | Proprietary | 1M | 69.0 | $30.00 |
| 25 | Hy4 | Tencent | Pending | 1M | 68.5 | $2.501 |
| 26 | GPT-5.4 Pro | OpenAI | Proprietary | 1.05M | 68.3 | $180.00 |
| 27 | Hy4 preview | Tencent | Open Weight | 1M | 68.2 | — |
| 28 | MiniMax M3.1 Flash | MiniMax | Proprietary | 1M | 68.1 | $0.60 |
| 29 | Gemini 3.8 Flash | Proprietary | 1M | 67.7 | $3.75 | |
| 30 | Ling 3.1 Flash | InclusionAI | Open Weight | 262K | 66.8 | — |
Reading the table
- Score — BenchLM’s BenchAlign composite across their benchmark suite. BenchLM never interpolates missing scores, so a model’s score can be lower-confidence without being wrong.
- Output $/1M — the provider’s list price per million output tokens. Input and cached prices are usually lower; check the model page before budgeting.
- Context — maximum context window.
Mis a million tokens. - — — the provider does not publish a list price (typically self-hosted or open weights).
Notes
- Ranked entries: 214. BenchLM separates Supported-ranked rows (exact source citations) from Estimated-ranked rows; lower ranks mix both.
- Scores move between snapshots as new evaluations land, so treat close ranks as ties.
- Regenerated automatically; the table above reflects the fetch on 2026-11-20.