Top benchmarked models and their cheapest blended price
Snapshot of the 25 highest-ranked models (Artificial Analysis Intelligence Index, 258 benchmarked models total). The interactive table above adds use-case tabs, live token-mix pricing, and capability-per-dollar value ranking.
| Model | Creator | AA Intelligence | AA Agentic | AA Coding | LiveBench Reasoning | Design Arena Elo | From $/M |
|---|---|---|---|---|---|---|---|
| anthropic/claude-sonnet-4-6 | anthropic | — | — | — | 84.8 | — | $3.06 |
| anthropic/claude-opus-5 | anthropic | 63.1 | 59.2 | 78 | — | 1382 | $0.735 |
| Claude Opus 5 (Fast) | anthropic | 63.1 | 59.2 | 78 | — | 1382 | $1.47 |
| Claude Opus 5 (batch) | anthropic | 63.1 | 59.2 | 78 | — | 1382 | $0.735 |
| anthropic/claude-fable-5 | anthropic | 62.1 | 56.6 | 76.5 | — | 1383 | $1.47 |
| Anthropic: Claude Fable 5 (batch) | anthropic | 62.1 | 56.6 | 76.5 | — | 1383 | $1.47 |
| gpt-5.6-sol | openai | 60.9 | 57.8 | 77.4 | — | — | $0.147 |
| SpaceXAI: Grok 4.6 | xai | 60.9 | 58.7 | 76.8 | — | 1340 | $0.565 |
| moonshotai/Kimi-K3 | moonshot | 59.7 | 54.3 | 76.2 | — | 1441 | $0.331 |
| Kimi K3 Fast | moonshot | 59.7 | 54.3 | 76.2 | — | 1441 | $0.441 |
| MoonshotAI: Kimi K3 (batch) | moonshot | 59.7 | 54.3 | 76.2 | — | 1441 | $0.376 |
| zai-org/GLM-5.3 | z-ai | 59.5 | 59.1 | 74.8 | — | — | $0.075 |
| Qwen/Qwen3.8-Max | qwen | 58.1 | 58.4 | 71.8 | — | 1368 | $0.266 |
| Qwen/Qwen3.8-2.4T-A95B | qwen | 57.7 | 57.1 | 71.9 | — | — | $0.274 |
| Qwen: Qwen3.8 2.4T A95B (batch) | qwen | 57.7 | 57.1 | 71.9 | — | — | $0.274 |
| zai-org/GLM-5.3-Flash | z-ai | 57.5 | 58.2 | 71.5 | — | — | $0.013 |
| Z.ai: GLM 5.3 Flash (batch) | z-ai | 57.5 | 58.2 | 71.5 | — | — | $0.016 |
| Anthropic: Claude Opus 4.8 (Fast) | anthropic | 57.3 | 49.4 | 74.3 | — | 1310 | $1.47 |
| Anthropic: Claude Opus 4.8 | anthropic | 57.3 | 49.4 | 74.3 | — | 1310 | $0.735 |
| Anthropic: Claude Opus 4.8 (batch) | anthropic | 57.3 | 49.4 | 74.3 | — | 1310 | $0.735 |
| Meta: Muse Spark 1.2 | — | 56.8 | 49.3 | 72.2 | — | 1357 | $0.198 |
| gpt-5.6-terra | openai | 56.6 | 50.2 | 76.7 | — | — | $0.152 |
| OpenAI: GPT-5.5 | openai | 56.3 | 47.4 | 74.9 | 89.7 | 1336 | $0.38 |
| google/gemini-3.7-flash | 56 | 45.1 | 76.1 | — | 1355 | $0.055 | |
| Google: Gemini 3.7 Flash (batch) | 56 | 45.1 | 76.1 | — | 1355 | $0.055 |
Scores from Artificial Analysis, LiveBench and Design Arena. "From $/M" is the cheapest provider's blended rate at a cached-heavy workload mix. See the benchmark FAQ for what each score measures.
Frequently asked questions
Where are the full FAQ lists?
All questions — text/token pricing, image and video generation pricing, and plain-language benchmark explainers — live on the consolidated FAQ page.