PRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.8010mPRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.0010mPRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.9810mPRICE~deepseek/deepseek-pro-latest input_per_mtok decreased from 0.13 to 0.1341mPRICE~deepseek/deepseek-pro-latest output_per_mtok increased from 0.40 to 4.2041mPRICE~deepseek/deepseek-pro-latest cached_input_per_mtok increased from 0.0042 to 0.1041mPRICE~moonshotai/kimi-latest input_per_mtok decreased from 0.99 to 0.7541mPRICE~moonshotai/kimi-latest output_per_mtok decreased from 13.00 to 11.3641mPRICE~moonshotai/kimi-latest cached_input_per_mtok decreased from 0.80 to 0.2941mPRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.991hPRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.001hPRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.801hPERFORMANCEGLM-5.2 → Scaleway: TTFT ↑ 20%2hPERFORMANCEDeepSeek V4 Flash → Scaleway: TTFT ↑ 35%2hPERFORMANCEMistral Medium 3.5 → Scaleway: TTFT ↓ 21%2hPERFORMANCEDeepSeek V4 Flash → Scaleway: throughput ↓ 20%2hPERFORMANCEGPT-OSS 120B → Scaleway: throughput ↑ 31%2hPERFORMANCEGPT-OSS 120B → Scaleway: TTFT ↓ 24%2hPERFORMANCEGPT-OSS 120B → Together AI: throughput ↑ 20%2hPERFORMANCEKimi K3 via OpenRouter: TTFT ↓ 49%2hPERFORMANCEGLM-5.2 via OpenRouter: reliability recovered 96.8% → 100.0%2hPERFORMANCEKimi K3 via Cortecs: throughput ↑ 184%2hPERFORMANCEKimi K3 via Cortecs: TTFT ↓ 42%2hPERFORMANCEGLM-5.2 via OpenRouter: throughput ↓ 49%2hPERFORMANCEGLM-5.2 via OpenRouter: TTFT ↑ 121%2hPERFORMANCEMiniMax M3 via OpenRouter: reliability recovered 96.8% → 100.0%2hPERFORMANCEDeepSeek V4 Pro via Cortecs: throughput ↑ 27%2hPERFORMANCEGPT-OSS 120B via OpenRouter: throughput ↓ 22%2hPERFORMANCEGLM-4.7 via OpenRouter: TTFT ↓ 78%2hPERFORMANCEGPT-OSS 20B → Groq: TTFT ↑ 24%2hPERFORMANCEKimi K3 → Together AI: TTFT ↓ 34%2hPERFORMANCEDeepSeek V4 Flash via OpenRouter: throughput ↓ 17%2hPERFORMANCELlama 3.3 70B via Cortecs: TTFT ↓ 51%2hPERFORMANCEQwen3 235B via OpenRouter: throughput ↑ 37%2hPERFORMANCEQwen3 235B via OpenRouter: TTFT ↓ 29%2hPERFORMANCEDeepSeek V4 Flash → Together AI: throughput ↑ 19%2hPERFORMANCEGLM-5.3 via OpenRouter: throughput ↓ 17%2hPERFORMANCEGPT-OSS 20B via Cortecs: throughput ↑ 28%2hPERFORMANCELlama 3.3 70B → Together AI: TTFT ↓ 33%2hPRICEtencent/hy4-preview output_per_mtok increased from 2.25 to 2.502h

Claude Sonnet 5

1 providers · — · 1M context · 1 regions · all models →
Not yet measured

Pricing comparisons publish with the first measured provider for Claude Sonnet 5.

Claude Sonnet 5 inference benchmark: fastest & cheapest providers · InferenceBench