Text, image, audio, and video combined
26
Models
25
Benchmarks
9
Providers
17
Open Weight
26
Deployable Now
17 self-host
26
Structured benchmark-backed rows
0
Signal-backed rows without full normalized tables
0
Benchmark-expected rows still catching up
| # | Model | Capability | Quality | Params | TAU-Bench | Aider Polyglot | MATH Lvl 5 | MMLU | $/M In | Speed | Open |
|---|---|---|---|---|---|---|---|---|---|---|---|
2 | Grok 4.6 xAI Subscription |
| 90.9 |
| 52.9 |
| 3.0T |
| 18.0 |
| 79.6 |
| 99.0 |
| 86.6 |
$2.00/M Cheapest verified |
| — |
71 | Kimi K2.7 Code Moonshot AI SubscriptionStructured | 87.6 | 72.6 | 1.0T | — | 59.1 | 97.1 | 82.4 | $0.6712/M Cheapest verified | — |
50 | Kimi K3 Moonshot AI Open WeightsStructured | 87.6 | 68.6 | 2.8T | 37.1 | — | — | — | $0.9900/M Cheapest verified | — |
24 | Qwen3.8 Max (0902) Qwen Open WeightsStructured | 86.3 | 31.7 | Unknown | — | — | — | — | $2.00/M Cheapest verified | — |
1 | Qwen3.8 27B Qwen Open WeightsStructured | 84.9 | 77.9 | 28B | — | — | — | — | $0.4200/M Cheapest verified | — |
45 | Claude Fable 5.1 Anthropic SubscriptionStructured | 81.8 | 61.4 | Undisclosed | — | — | — | — | $10.00/M Cheapest verified | — |
62 | Kimi K2.5 Moonshot AI SubscriptionStructured | 80.1 | 65.6 | 1.0T | — | 59.1 | — | — | $0.4500/M Cheapest verified | — |
142 | Qwen3.5-4B Qwen Open WeightsStructured | 79.7 | 62.5 | 5B | 79.9 | — | — | 88.8 | Free Free | — |
27 | Qwen3.6 Plus Qwen Open WeightsStructured | 79.5 | 28.7 | Unknown | — | — | — | — | $0.3250/M Cheapest verified | — |
67 | Kimi K2.6 Moonshot AI Open WeightsStructured | 79.1 | 73.1 | 1.0T | — | — | — | — | $0.9500/M Cheapest verified | — |
173 | diffusiongemma-26B-A4B-it | 77.5 | 59.1 | 26B | — | — | — | 81.5 | Free Free | — |
54 | Claude Fable 5 Anthropic SubscriptionStructured | 76.5 | 61.8 | Undisclosed | 39.7 | — | — | — | $10.00/M Cheapest verified | — |
53 | Claude Sonnet 5 Anthropic SubscriptionStructured | 76.3 | 63.6 | Undisclosed | — | — | — | — | $2.00/M Cheapest verified | — |
51 | Claude Opus 5 Anthropic SubscriptionStructured | 74.8 | 57.8 | Undisclosed | 48.7 | — | — | — | $4.00/M Cheapest verified | — |
48 | Claude 4 Sonnet Anthropic SubscriptionStructured | 74.7 | 42.7 | Undisclosed | 63.5 | — | — | 90.4 | $3.00/M Cheapest verified | 95 tok/s |
167 | gemma-4-31B-it-qat-w4a16-ct | 74.2 | 55.6 | 31B | — | — | — | 88.4 | Free Free | — |
69 | GPT-4o OpenAI SubscriptionStructured | 73.7 | 41.3 | Undisclosed | 55.8 | 45.3 | 78.9 | 88.7 | $2.50/M Cheapest verified | 109 tok/s |
22 | Muse Glimmer 30B Meta Open WeightsStructured | 73.6 | 30.7 | 30B | — | — | — | — | $0.3500/M Cheapest verified | — |
179 | gemma-4-26B-A4B-it-qat-q4_0-unquantized | 73.5 | 54.2 | 27B | — | — | — | 88.4 | Free Free | — |
5 | Gemma 4 26B A4B IT | 72.8 | 71.7 | 26B | — | — | — | 86.3 | $0.0675/M Cheapest verified | — |
149 | DeepSeek-OCR DeepSeek Open WeightsStructured | 72.3 | 51.7 | 3B | — | 74.2 | — | — | Free Free | — |
21 | Llama 4 Maverick Meta Open WeightsStructured | 71.2 | 40.9 | 400B | 45.7 | 15.6 | 88.9 | 80.9 | Free Cheapest verified | — |
172 | DeepSeek-OCR-2 DeepSeek Open WeightsStructured | 70.2 | 51.8 | 3B | — | 74.2 | — | — | Free Free | — |
34 | Qwen3.5 Plus 2026-02-15 Qwen Open WeightsStructured | 69.8 | 26.3 | Unknown | — | 8.0 | — | — | $0.2600/M Cheapest verified | — |
25 | Mistral Large 3 Mistral AI Open WeightsStructured | 69.4 | 40.1 | 675B | — | 11.1 | — | 80.7 | $0.5000/M Cheapest verified | — |
72 | Qwen2.5 VL 72B Instruct Qwen Open WeightsStructured | 54.4 | 49.4 | 73B | — | 8.0 | — | — | $0.8000/M Cheapest verified | — |