PRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.8031m PRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.0031m PRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.9931m PERFORMANCEGLM-5.2 → Scaleway: TTFT ↑ 20%1h PERFORMANCEDeepSeek V4 Flash → Scaleway: TTFT ↑ 35%1h PERFORMANCEMistral Medium 3.5 → Scaleway: TTFT ↓ 21%1h PERFORMANCEDeepSeek V4 Flash → Scaleway: throughput ↓ 20%1h PERFORMANCEGPT-OSS 120B → Scaleway: throughput ↑ 31%1h PERFORMANCEGPT-OSS 120B → Scaleway: TTFT ↓ 24%1h PERFORMANCEGPT-OSS 120B → Together AI: throughput ↑ 20%1h PERFORMANCEKimi K3 via OpenRouter: TTFT ↓ 49%1h PERFORMANCEGLM-5.2 via OpenRouter: reliability recovered 96.8% → 100.0%1h PERFORMANCEKimi K3 via Cortecs: throughput ↑ 184%1h PERFORMANCEKimi K3 via Cortecs: TTFT ↓ 42%1h PERFORMANCEGLM-5.2 via OpenRouter: throughput ↓ 49%1h PERFORMANCEGLM-5.2 via OpenRouter: TTFT ↑ 121%1h PERFORMANCEMiniMax M3 via OpenRouter: reliability recovered 96.8% → 100.0%1h PERFORMANCEDeepSeek V4 Pro via Cortecs: throughput ↑ 27%1h PERFORMANCEGPT-OSS 120B via OpenRouter: throughput ↓ 22%1h PERFORMANCEGLM-4.7 via OpenRouter: TTFT ↓ 78%1h PERFORMANCEGPT-OSS 20B → Groq: TTFT ↑ 24%1h PERFORMANCEKimi K3 → Together AI: TTFT ↓ 34%1h PERFORMANCEDeepSeek V4 Flash via OpenRouter: throughput ↓ 17%1h PERFORMANCELlama 3.3 70B via Cortecs: TTFT ↓ 51%1h PERFORMANCEQwen3 235B via OpenRouter: throughput ↑ 37%1h PERFORMANCEQwen3 235B via OpenRouter: TTFT ↓ 29%1h PERFORMANCEDeepSeek V4 Flash → Together AI: throughput ↑ 19%1h PERFORMANCEGLM-5.3 via OpenRouter: throughput ↓ 17%1h PERFORMANCEGPT-OSS 20B via Cortecs: throughput ↑ 28%1h PERFORMANCELlama 3.3 70B → Together AI: TTFT ↓ 33%1h PRICE~moonshotai/kimi-latest cached_input_per_mtok decreased from 0.80 to 0.291h PRICE~moonshotai/kimi-latest output_per_mtok decreased from 13.00 to 11.361h PRICE~moonshotai/kimi-latest input_per_mtok decreased from 0.99 to 0.751h PRICE~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.0077 to 0.00111h PRICE~deepseek/deepseek-v4-flash-latest output_per_mtok decreased from 1.28 to 1.041h PRICE~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.0077 to 0.00381h PRICEz-ai/glm-5.1 cached_input_per_mtok increased from 0.18 to 0.261h PRICEz-ai/glm-5.1 output_per_mtok increased from 3.03 to 4.401h PRICEz-ai/glm-5.1 input_per_mtok increased from 0.96 to 1.401h PRICEtencent/hy4-preview cached_input_per_mtok increased from 0.038 to 0.0421h

PRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.8031m PRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.0031m PRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.9931m PERFORMANCEGLM-5.2 → Scaleway: TTFT ↑ 20%1h PERFORMANCEDeepSeek V4 Flash → Scaleway: TTFT ↑ 35%1h PERFORMANCEMistral Medium 3.5 → Scaleway: TTFT ↓ 21%1h PERFORMANCEDeepSeek V4 Flash → Scaleway: throughput ↓ 20%1h PERFORMANCEGPT-OSS 120B → Scaleway: throughput ↑ 31%1h PERFORMANCEGPT-OSS 120B → Scaleway: TTFT ↓ 24%1h PERFORMANCEGPT-OSS 120B → Together AI: throughput ↑ 20%1h PERFORMANCEKimi K3 via OpenRouter: TTFT ↓ 49%1h PERFORMANCEGLM-5.2 via OpenRouter: reliability recovered 96.8% → 100.0%1h PERFORMANCEKimi K3 via Cortecs: throughput ↑ 184%1h PERFORMANCEKimi K3 via Cortecs: TTFT ↓ 42%1h PERFORMANCEGLM-5.2 via OpenRouter: throughput ↓ 49%1h PERFORMANCEGLM-5.2 via OpenRouter: TTFT ↑ 121%1h PERFORMANCEMiniMax M3 via OpenRouter: reliability recovered 96.8% → 100.0%1h PERFORMANCEDeepSeek V4 Pro via Cortecs: throughput ↑ 27%1h PERFORMANCEGPT-OSS 120B via OpenRouter: throughput ↓ 22%1h PERFORMANCEGLM-4.7 via OpenRouter: TTFT ↓ 78%1h PERFORMANCEGPT-OSS 20B → Groq: TTFT ↑ 24%1h PERFORMANCEKimi K3 → Together AI: TTFT ↓ 34%1h PERFORMANCEDeepSeek V4 Flash via OpenRouter: throughput ↓ 17%1h PERFORMANCELlama 3.3 70B via Cortecs: TTFT ↓ 51%1h PERFORMANCEQwen3 235B via OpenRouter: throughput ↑ 37%1h PERFORMANCEQwen3 235B via OpenRouter: TTFT ↓ 29%1h PERFORMANCEDeepSeek V4 Flash → Together AI: throughput ↑ 19%1h PERFORMANCEGLM-5.3 via OpenRouter: throughput ↓ 17%1h PERFORMANCEGPT-OSS 20B via Cortecs: throughput ↑ 28%1h PERFORMANCELlama 3.3 70B → Together AI: TTFT ↓ 33%1h PRICE~moonshotai/kimi-latest cached_input_per_mtok decreased from 0.80 to 0.291h PRICE~moonshotai/kimi-latest output_per_mtok decreased from 13.00 to 11.361h PRICE~moonshotai/kimi-latest input_per_mtok decreased from 0.99 to 0.751h PRICE~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.0077 to 0.00111h PRICE~deepseek/deepseek-v4-flash-latest output_per_mtok decreased from 1.28 to 1.041h PRICE~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.0077 to 0.00381h PRICEz-ai/glm-5.1 cached_input_per_mtok increased from 0.18 to 0.261h PRICEz-ai/glm-5.1 output_per_mtok increased from 3.03 to 4.401h PRICEz-ai/glm-5.1 input_per_mtok increased from 0.96 to 1.401h PRICEtencent/hy4-preview cached_input_per_mtok increased from 0.038 to 0.0421h

GPT-OSS 20B

6 providers · 21B MoE · 128k context · 2 regions · all models →

GPT-OSS 20B provider ranking

#1 OverallGroq99.5 FastestGroq0.69 s Best valueOpenRouter Europe #1Cortecs82.3 Best toolsGroq100.0%Best structuredGroq100.0%

#ProviderΔ30d

01GroqBEST OVERALL● MEASUREDUS · Serverless · GPT-OSS 20B100.0%100.0%100.0%2583680.7s929P33↑ 0.4

02CortecsBEST EUROPEANEU RESIDENT● MEASUREDAT · Routed · GPT-OSS 20B99.5%100.0%100.0%1755886.2s67P67↑ 0.0

03OpenRouterBEST VALUE● MEASUREDUS · Routed · GPT-OSS 20B100.0%100.0%95.2%65618966.0s75P100↓ 0.1

3 of 6 providers ranked · · Methodology →

Sort

01Groq99.5

Best overallUS · Serverless · GPT-OSS 20B

TTFT258 ms

Rel.100.0%

Tok/s929

$ Out$0.30

Tools100.0%

Δ30d+0.4

Best EuropeanAT · Routed · GPT-OSS 20B

TTFT175 ms

Rel.99.5%

Tok/s67

$ Out$0.17

Tools100.0%

Δ30d0.0

03OpenRouter79.3

Best valueUS · Routed · GPT-OSS 20B

TTFT656 ms

Rel.100.0%

Tok/s75

$ Out$0.09

Tools100.0%

Δ30d-0.1

3 of 6 providers ranked · · Methodology →

GPT-OSS 20B inference benchmark: fastest & cheapest providers · InferenceBench