PRICE~z-ai/glm-latest cached_input_per_mtok increased from 0.08 to 0.1519m PRICE~z-ai/glm-latest output_per_mtok decreased from 4.00 to 2.7919m PRICE~z-ai/glm-latest input_per_mtok decreased from 0.12 to 0.06619m PRICE~moonshotai/kimi-latest cached_input_per_mtok decreased from 0.80 to 0.4519m PRICE~moonshotai/kimi-latest input_per_mtok decreased from 0.98 to 0.9719m PRICE~deepseek/deepseek-v4-flash-latest output_per_mtok decreased from 1.04 to 0.8119m PRICE~deepseek/deepseek-v4-flash-latest input_per_mtok increased from 0.0038 to 0.008619m PRICE~deepseek/deepseek-pro-latest cached_input_per_mtok decreased from 0.10 to 0.004219m PRICE~deepseek/deepseek-pro-latest output_per_mtok decreased from 4.20 to 0.4019m PRICE~deepseek/deepseek-pro-latest input_per_mtok increased from 0.13 to 0.1319m PRICEdeepseek/deepseek-v4-flash-0731 cached_input_per_mtok increased from 0.0051 to 0.01219m PRICEdeepseek/deepseek-v4-flash-0731 input_per_mtok increased from 0.0051 to 0.01219m PRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.8050m PRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.0050m PRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.9850m PRICE~deepseek/deepseek-pro-latest input_per_mtok decreased from 0.13 to 0.131h PRICE~deepseek/deepseek-pro-latest output_per_mtok increased from 0.40 to 4.201h PRICE~deepseek/deepseek-pro-latest cached_input_per_mtok increased from 0.0042 to 0.101h PRICE~moonshotai/kimi-latest input_per_mtok decreased from 0.99 to 0.751h PRICE~moonshotai/kimi-latest output_per_mtok decreased from 13.00 to 11.361h PRICE~moonshotai/kimi-latest cached_input_per_mtok decreased from 0.80 to 0.291h PRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.992h PRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.002h PRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.802h PERFORMANCEGLM-5.2 → Scaleway: TTFT ↑ 20%2h PERFORMANCEDeepSeek V4 Flash → Scaleway: TTFT ↑ 35%2h PERFORMANCEMistral Medium 3.5 → Scaleway: TTFT ↓ 21%2h PERFORMANCEDeepSeek V4 Flash → Scaleway: throughput ↓ 20%2h PERFORMANCEGPT-OSS 120B → Scaleway: throughput ↑ 31%2h PERFORMANCEGPT-OSS 120B → Scaleway: TTFT ↓ 24%2h PERFORMANCEGPT-OSS 120B → Together AI: throughput ↑ 20%2h PERFORMANCEKimi K3 via OpenRouter: TTFT ↓ 49%2h PERFORMANCEGLM-5.2 via OpenRouter: reliability recovered 96.8% → 100.0%2h PERFORMANCEKimi K3 via Cortecs: throughput ↑ 184%2h PERFORMANCEKimi K3 via Cortecs: TTFT ↓ 42%2h PERFORMANCEGLM-5.2 via OpenRouter: throughput ↓ 49%2h PERFORMANCEGLM-5.2 via OpenRouter: TTFT ↑ 121%2h PERFORMANCEMiniMax M3 via OpenRouter: reliability recovered 96.8% → 100.0%2h PERFORMANCEDeepSeek V4 Pro via Cortecs: throughput ↑ 27%2h PERFORMANCEGPT-OSS 120B via OpenRouter: throughput ↓ 22%2h

PRICE~z-ai/glm-latest cached_input_per_mtok increased from 0.08 to 0.1519m PRICE~z-ai/glm-latest output_per_mtok decreased from 4.00 to 2.7919m PRICE~z-ai/glm-latest input_per_mtok decreased from 0.12 to 0.06619m PRICE~moonshotai/kimi-latest cached_input_per_mtok decreased from 0.80 to 0.4519m PRICE~moonshotai/kimi-latest input_per_mtok decreased from 0.98 to 0.9719m PRICE~deepseek/deepseek-v4-flash-latest output_per_mtok decreased from 1.04 to 0.8119m PRICE~deepseek/deepseek-v4-flash-latest input_per_mtok increased from 0.0038 to 0.008619m PRICE~deepseek/deepseek-pro-latest cached_input_per_mtok decreased from 0.10 to 0.004219m PRICE~deepseek/deepseek-pro-latest output_per_mtok decreased from 4.20 to 0.4019m PRICE~deepseek/deepseek-pro-latest input_per_mtok increased from 0.13 to 0.1319m PRICEdeepseek/deepseek-v4-flash-0731 cached_input_per_mtok increased from 0.0051 to 0.01219m PRICEdeepseek/deepseek-v4-flash-0731 input_per_mtok increased from 0.0051 to 0.01219m PRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.8050m PRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.0050m PRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.9850m PRICE~deepseek/deepseek-pro-latest input_per_mtok decreased from 0.13 to 0.131h PRICE~deepseek/deepseek-pro-latest output_per_mtok increased from 0.40 to 4.201h PRICE~deepseek/deepseek-pro-latest cached_input_per_mtok increased from 0.0042 to 0.101h PRICE~moonshotai/kimi-latest input_per_mtok decreased from 0.99 to 0.751h PRICE~moonshotai/kimi-latest output_per_mtok decreased from 13.00 to 11.361h PRICE~moonshotai/kimi-latest cached_input_per_mtok decreased from 0.80 to 0.291h PRICE~moonshotai/kimi-latest input_per_mtok increased from 0.75 to 0.992h PRICE~moonshotai/kimi-latest output_per_mtok increased from 11.36 to 13.002h PRICE~moonshotai/kimi-latest cached_input_per_mtok increased from 0.29 to 0.802h PERFORMANCEGLM-5.2 → Scaleway: TTFT ↑ 20%2h PERFORMANCEDeepSeek V4 Flash → Scaleway: TTFT ↑ 35%2h PERFORMANCEMistral Medium 3.5 → Scaleway: TTFT ↓ 21%2h PERFORMANCEDeepSeek V4 Flash → Scaleway: throughput ↓ 20%2h PERFORMANCEGPT-OSS 120B → Scaleway: throughput ↑ 31%2h PERFORMANCEGPT-OSS 120B → Scaleway: TTFT ↓ 24%2h PERFORMANCEGPT-OSS 120B → Together AI: throughput ↑ 20%2h PERFORMANCEKimi K3 via OpenRouter: TTFT ↓ 49%2h PERFORMANCEGLM-5.2 via OpenRouter: reliability recovered 96.8% → 100.0%2h PERFORMANCEKimi K3 via Cortecs: throughput ↑ 184%2h PERFORMANCEKimi K3 via Cortecs: TTFT ↓ 42%2h PERFORMANCEGLM-5.2 via OpenRouter: throughput ↓ 49%2h PERFORMANCEGLM-5.2 via OpenRouter: TTFT ↑ 121%2h PERFORMANCEMiniMax M3 via OpenRouter: reliability recovered 96.8% → 100.0%2h PERFORMANCEDeepSeek V4 Pro via Cortecs: throughput ↑ 27%2h PERFORMANCEGPT-OSS 120B via OpenRouter: throughput ↓ 22%2h

GPT-OSS 120B

11 providers · 117B MoE · 128k context · 2 regions · all models →

GPT-OSS 120B provider ranking

#1 OverallCerebras99.0 FastestCerebras0.41 s Best valueOpenRouter Europe #1Cortecs97.6 Best toolsCerebras100.0%Best structuredCerebras100.0%

#ProviderΔ30d

01CerebrasBEST OVERALL● MEASUREDUS · Serverless · GPT-OSS 120B100.0%100.0%100.0%1843230.4s1749P33↑ 0.0

02Groq● MEASUREDUS · Serverless · GPT-OSS 120B100.0%100.0%100.0%2643991.1s486P67↑ 0.8

03CortecsBEST EUROPEANEU RESIDENT● MEASUREDAT · Routed · GPT-OSS 120B100.0%100.0%100.0%2375211.4s351P83↓ 1.3

04ScalewayEU RESIDENTEU OWNED● MEASUREDFR · Serverless · GPT-OSS 120B100.0%100.0%100.0%17922144.4s96P50↑ 0.2

05OpenRouterBEST VALUE● MEASUREDUS · Routed · GPT-OSS 120B99.1%100.0%100.0%47414136.4s68P100↑ 7.0

06Together● MEASUREDUS · Serverless · GPT-OSS 120B93.5%100.0%85.7%1824673.1s138P17↑ 14.9

6 of 11 providers ranked · · Methodology →

Sort

Best overallUS · Serverless · GPT-OSS 120B

TTFT184 ms

Rel.100.0%

Tok/s1749

$ Out$0.75

Tools100.0%

Δ30d0.0

02Groq97.8

US · Serverless · GPT-OSS 120B

TTFT264 ms

Rel.100.0%

Tok/s486

$ Out$0.60

Tools100.0%

Δ30d+0.8

Best EuropeanAT · Routed · GPT-OSS 120B

TTFT237 ms

Rel.100.0%

Tok/s351

$ Out$0.46

Tools100.0%

Δ30d-1.3

FR · Serverless · GPT-OSS 120B

TTFT179 ms

Rel.100.0%

Tok/s96

$ Out$0.69

Tools100.0%

Δ30d+0.2

05OpenRouter82.4

Best valueUS · Routed · GPT-OSS 120B

TTFT474 ms

Rel.99.1%

Tok/s68

$ Out$0.17

Tools100.0%

Δ30d+7.0

US · Serverless · GPT-OSS 120B

TTFT182 ms

Rel.93.5%

Tok/s138

$ Out$0.60

Tools100.0%

Δ30d+14.9

6 of 11 providers ranked · · Methodology →

GPT-OSS 120B inference benchmark: fastest & cheapest providers · InferenceBench