Fastest
deepseek-v4.1-flash-1017
64.1токенов/с
1.8 × faster than the catalog average
Lowest latency
gemini-3.8-flash
9.69s
Catalog average latency: 15.81 s
Highest savings
gpt-5.4
−98%
Discount vs. the provider's official rate
Click a column header to sort
| # | Model | |||||||
|---|---|---|---|---|---|---|---|---|
| 1 | claude-opus-5-5Anthropic | 524,043 | 45.3 B | $1.84 | −21 % | 84.1 % | 17.26 s | 49.4 t/s |
| 2 | gpt-6-astraOpenAI | 521,478 | 43.2 B | $99.74 | −88 % | 81.9 % | 17.14 s | 50.3 t/s |
| 3 | claude-sonnet-5Anthropic | 294,168 | 16.2 B | $0.15 | −94 % | 93.3 % | 10.93 s | 27.4 t/s |
| 4 | deepseek-v4-flashDeepSeek | 111,319 | 1.7 B | $0.0038 | — | 42.9 % | 18.44 s | 18.9 t/s |
| 5 | claude-opus-5Anthropic | 100,827 | 7.8 B | $0.66 | −92 % | 89.3 % | 13.51 s | 59.8 t/s |
| 6 | claude-fable-5-1Anthropic | 100,013 | 8.1 B | $0.0026 | −85 % | 83.1 % | 20.48 s | 56.8 t/s |
| 7 | gemini-3.8-flashGoogle | 68,650 | 751.5 M | $0 | −95 % | 84.7 % | 9.69 s | 31 t/s |
| 8 | claude-haiku-4-5-20251001Anthropic | 56,095 | 1.2 B | $0.001 | −98 % | 66.1 % | 22.48 s | 23.6 t/s |
| 9 | gpt-5.6-solOpenAI | 51,413 | 4 B | $0.79 | −84 % | 81.3 % | 16.23 s | 38.5 t/s |
| 10 | gpt-5.6-lunaOpenAI | 49,298 | 2.7 B | $0.42 | −69 % | 73.2 % | 16.24 s | 26.3 t/s |
| 11 | deepseek-v4.1-flash-1017DeepSeek | 36,916 | 4.9 B | $0.000074 | — | 94.8 % | 12.32 s | 64.1 t/s |
| 12 | gpt-5.6-terraOpenAI | 26,399 | 1.8 B | $0.00063 | −91 % | 86.6 % | 20.02 s | 45.7 t/s |
| 13 | claude-fable-5Anthropic | 24,226 | 2.6 B | $0 | −81 % | 92.1 % | 14.98 s | 40 t/s |
| 14 | deepseek-v4-proDeepSeek | 23,066 | 1.4 B | $0.0046 | — | 76.5 % | 19.84 s | 30.6 t/s |
| 15 | claude-opus-4-8Anthropic | 13,928 | 844.2 M | $0.000052 | −96 % | 88.2 % | 17.19 s | 27.2 t/s |
| 16 | codex-auto-reviewOpenAI | 11,454 | 770.6 M | $2.55 | −93 % | 91.2 % | 10.95 s | 15.4 t/s |
| 17 | glm-5.3Z.AI | 6,675 | 669.9 M | $0.0007 | — | 87.1 % | 16.18 s | 45.2 t/s |
| 18 | gemini-3.7-flashGoogle | 6,309 | 263.4 M | $0.00012 | −95 % | 58.1 % | 13.45 s | 40.5 t/s |
| 19 | gpt-5.4-miniOpenAI | 6,050 | 17.1 M | $0.00032 | −92 % | 43.7 % | 18.28 s | 17.4 t/s |
| 20 | gpt-5.5OpenAI | 4,120 | 87 M | $0.0019 | −95 % | 62.5 % | 15.42 s | 29.7 t/s |
| 21 | gpt-5.4OpenAI | 2,761 | 124.6 M | $0.000084 | −98 % | 80.5 % | 15.31 s | 34.4 t/s |
| 22 | grok-4.70xAI | 2,649 | 47.4 M | $0 | −89 % | 58 % | 13.7 s | 37.6 t/s |
| 23 | gpt-oss-120bOpenAI | 1,175 | 14.7 M | $0 | — | 78.8 % | 13.58 s | 28.5 t/s |
| 24 | gpt-image-2.5OpenAI | 977 | — | $0 | −95 % | 77.5 % | 38.02 s | — |
| 25 | gemini-3.1-flash-lite-imageGoogle | 667 | — | $0 | −95 % | 0 % | — | — |
| 26 | gpt-image-2OpenAI | 394 | — | $0 | −95 % | 27.7 % | 41.3 s | — |
Usage statistics update as API requests come in.uptime 80 % and above50–80 %below 50 %