GPT-4.1 Mini

OpenAI · 1M input + 1M output tokens · USD

Choose a model

Prices use one fixed workload so the gateway totals are directly comparable.

ModelGPT-4.1 MiniOpenAI

 Claude Fable 5 AnthropicClaude Fable 5.1 anthropicClaude Haiku 4.5 AnthropicClaude Haiku 5.5 anthropicClaude Opus 4.5 AnthropicClaude Opus 4.6 AnthropicClaude Opus 4.7 AnthropicClaude Opus 4.8 AnthropicClaude Opus 5 AnthropicClaude Opus 5.5 anthropicClaude Sonnet 4.5 AnthropicClaude Sonnet 4.6 AnthropicClaude Sonnet 5 AnthropicClaude Sonnet 5.5 anthropicCommand A cohereDeepSeek V3.1 Terminus deepseekDeepSeek V3.2 deepseekDeepSeek V4 Flash deepseekDeepSeek V4 Flash Vision Exp deepseekDeepSeek V4 Pro 0423 deepseekDeepSeek V4.1 Flash deepseekDeepSeek V4.1 Flash 1017 deepseekdeepseek-v4-flash-0731 deepseek-aideepseek-v4-pro-0813 deepseek-aiEmber-1 fireworksFugu Max sakanaFugu Ultra sakanaFugu Ultra v2 sakanaGemini 2.5 Flash GoogleGemini 2.5 Flash Lite googleGemini 2.5 Pro googleGemini 3.1 Flash Lite googleGemini 3.1 Pro Preview googleGemini 3.5 Flash googleGemini 3.5 Flash Lite googleGemini 3.6 Flash googleGemini 3.7 Flash googleGemini 3.8 Flash googleGemini 4 Argon googleGemma 4 31B googlegemma-4-26b-a4b-it googleGlimpse elorianGLM 5 Turbo z-aiGLM 5.3 FlashX z-aiGLM 5V Turbo z-aiGLM-4.5 Z.aiGLM-4.5 Air Z.aiGLM-4.5V Z.aiGLM-4.6 Z.aiGLM-4.7 Z.aiGLM-4.7 Flash Z.aiGLM-5 Z.aiGLM-5.1 Z.aiGLM-5.2 Z.aiglm-5.3 zai-orgglm-5.3-flash zai-orgGPT-3.5 Turbo openaiGPT-4 Turbo openaiGPT-4.1 OpenAIGPT-4.1 Mini OpenAIGPT-4.1 Nano OpenAIGPT-4o OpenAIGPT-4o Mini OpenAIGPT-5 OpenAIGPT-5 Mini OpenAIGPT-5 Nano OpenAIGPT-5 Pro OpenAIGPT-5.1-Codex openaiGPT-5.1-Codex-Max openaiGPT-5.1-Codex-Mini openaiGPT-5.2 OpenAIGPT-5.2 Pro OpenAIGPT-5.2-Codex openaiGPT-5.3-Codex openaiGPT-5.4 OpenAIGPT-5.4 Mini OpenAIGPT-5.4 Nano OpenAIGPT-5.4 Pro OpenAIGPT-5.5 OpenAIGPT-5.5 Pro OpenAIGPT-5.6 Luna OpenAIGPT-5.6 Sol OpenAIGPT-5.6 Terra OpenAIGPT-6 Astra OpenAIGPT-6 Luna openaiGPT-6 Sol openaiGPT-6.1 Sol openaigpt-oss-120b openaigpt-oss-20b openaigpt-oss-safeguard-20b openaigranite-4.0-h-micro ibm-graniteHunyuan A13B Instruct tencentHy3 tencentHy4 preview tencentInkling thinkingmachinesInkling Small thinkingmachinesKimi K2 0711 moonshotaiKimi K2.5 moonshotaiKimi K3 moonshotaikimi-k2.6 moonshotaikimi-k2.7-code moonshotaiLaguna S 2.1 poolsideLing 3.0 Flash inclusionaiLing 3.0 Flash VL inclusionaiLing 3.1 Flash inclusionaiLlama 4 Maverick meta-llamaLlama 4 Scout meta-llamallama-3.2-1b-instruct metallama-3.2-3b-instruct metaLongCat 2.0 meituanMercury 2 inceptionMercury 2.5 inceptionMiMo-V2.5 xiaomiMiMo-V2.5-Pro xiaomiMiMo-V2.6-Flash xiaomiMiMo-V2.6-Pro xiaomiMiMo-V2.6-Pro-UltraSpeed xiaomiMiniMax M2 minimaxMiniMax M2.1 minimaxMiniMax M2.5 minimaxMiniMax M2.7 minimaxMiniMax M3 minimaxMistral Nemo mistralaimistral-small-3.1-24b-instruct mistralaimoondream3.1-9B-A2B moondreamMorph V3 Fast morphMorph V3 Large morphMuse Glimmer 30B metaMuse Spark 1.1 metaMuse Spark 1.2 metaMuse Spark 1.2 Contributor metaMuse Spark 1.3 metaMuse Spark 1.3 Contributor metaMuse Spark Open metaNano Banana (Gemini 2.5 Flash Image) googleNano Banana 2 (Gemini 3.1 Flash Image Preview) googleNano Banana 2 (Gemini 3.1 Flash Image) googleNano Banana 2 Lite (Gemini 3.1 Flash Lite Image) googleNano Banana 2.1 googleNano Banana Pro (Gemini 3 Pro Image) googleNemotron 3 Nano 30B A3B nvidiaNemotron 3 Super nvidiaNemotron 3 Ultra nvidiaNemotron 3.5 Lightning nvidiaNemotron 3.5 Super VL nvidiao1 OpenAIo3 OpenAIo3 Mini OpenAIo3 Pro OpenAIo4 Mini OpenAIQwen3 14B qwenQwen3 30B A3B Instruct 2507 qwenQwen3 32B qwenQwen3 Coder 30B A3B Instruct qwenQwen3 Coder 480B A35B qwenQwen3 Coder Next qwenQwen3 Next 80B A3B Instruct qwenQwen3 Next 80B A3B Thinking qwenQwen3 VL 30B A3B Instruct qwenQwen3 VL 30B A3B Thinking qwenQwen3 VL 8B Instruct qwenQwen3.5-122B-A10B qwenQwen3.5-27B qwenQwen3.5-35B-A3B qwenQwen3.5-9B qwenQwen3.6 27B qwenQwen3.6 35B A3B qwenQwen3.6 Plus qwenQwen3.7 Flash qwenQwen3.7 Max qwenQwen3.7 Plus qwenQwen3.8 2.4T A95B qwenQwen3.8 Flash qwenQwen3.8 Max (0902) qwenQwen3.8 Max Prime qwenQwen3.8 Omni Flash qwenqwen3.8-27b qwenR1 deepseekSchematron V2 Small inference-netSchematron V2 Turbo inference-netSeed 1.6 bytedance-seedSonar perplexityStep 3.5 Flash stepfunStep 3.7 Flash stepfunStep 5 Preview stepfunTrinity Large Thinking arcee-ai

Same model · same workloadGPT-4.1 MiniOfficial API baseline$2

OpenRouter1M input + 1M output$2.11Model usage$2Credit purchase fee estimate+$0.115.5% when purchasing credits, not per requestVercel AI GatewayLowest gateway cost1M input + 1M output$2Model usage$2Gateway markup$0No token markupCloudflare AI Gateway1M input + 1M output$2.1Model usage$2Unified Billing fee estimate+$0.15% unified billing feePortkey1M input + 1M output$2 + planModel usage$2Platform planSeparate planProvider usage plus a separate platform plan

Estimated total uses 1 million input tokens plus 1 million output tokens. OpenRouter's funding estimate applies 5.5% to this amount for comparison; the real charge occurs when credits are purchased and has a $0.80 minimum per purchase. Taxes, payment charges, discounts, and monthly plans are excluded.