10,793 prices on 4,759 listings · last collected 12 min ago · 403 flagged changes in 30 days
Every price, as each platform publishes it
A collector reads every public price on eight platforms, daily by default, and keeps the page it read. Nothing is typed in by hand. Search a model to see who charges least, per unit, right now.
Model
Claude 
Gemini 
Qwen 
DeepSeek 
GLM 
Kimi 
Mistral 
MiniMax 
Hunyuan 
Nemotron 
Llama Unit price
All 3,235Per second of output 13Per image 6Per million input tokens 1,107Per million output tokens 1,071Per million tokens 5Per million cached input tokens (read) 594Per million cache-write tokens 224Per thousand characters 2Per second of compute time 12Per run (default parameters) 193Other units (see the platform’s wording) 8
Showing 301–400 of 594
Per million cached input tokens (read)100
| Platform | Model · what it covers | USD | Platform’s own wording | Since |
|---|---|---|---|---|
| OpenRouterOfficial | OpenAI: GPT-5.6 Terra (batch) | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint openai) | |
| OpenRouterOfficial | OpenAI: GPT-6 Sol Pro (batch) | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint openai) | |
| OpenRouterOfficial | OpenAI: GPT-6 Sol (batch) | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint openai) | |
| OpenRouterOfficial partner | OpenAI: GPT-6.1 Solvia Amazon Bedrock · global | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint amazon-bedrock/global) | |
| OpenRouterOfficial partner | OpenAI: GPT-6.1 Sol Provia Azure | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint azure) | |
| OpenRouterOfficial | OpenAI: GPT-6.1 Sol Pro (batch)Prompt > 272K | $0.1/M cache read | input_cache_read: 0.0000001 / token (>272000; endpoint openai) | |
| OpenRouterOfficial | OpenAI: GPT-6.1 Sol (batch)Prompt > 272K | $0.1/M cache read | input_cache_read: 0.0000001 / token (>272000; endpoint openai) | |
| OpenRouterOfficial partner | OpenAI: GPT-6.1 TerraPrompt > 272Kvia Azure | $0.1/M cache read | input_cache_read: 0.0000001 / token (>272000; endpoint azure) | |
| OpenRouterOfficial partner | OpenAI: GPT-6.1 Terra ProPrompt > 272Kvia Azure | $0.1/M cache read | input_cache_read: 0.0000001 / token (>272000; endpoint azure) | |
| OpenRouterThird-party | Qwen: Qwen3 Coder 480B A35BTurbovia DeepInfra | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint deepinfra/turbo) | |
| OpenRouterThird-party | StepFun: Step 5 Previewvia StepFun · fp8 | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint stepfun/fp8) | |
| OpenRouterThird-party | Thinking Machines: Inkling Small | $0.1/M cache read | input_cache_read: 0.0000001 / token (endpoint deepinfra/fp8) | |
| OpenRouterNot disclosed | Anthropic: Claude Sonnet Latest | $0.1/M cache read | input_cache_read: 0.0000001 / token | |
| OpenRouterNot disclosed | Google: Gemini Pro Latest | $0.1/M cache read | input_audio_cache: 0.0000001 / token | |
| OpenRouterNot disclosed | Google: Gemini Pro Latest | $0.1/M cache read | input_cache_read: 0.0000001 / token | |
| OpenRouterNot disclosed | OpenAI: GPT Sol Latest | $0.1/M cache read | input_cache_read: 0.0000001 / token | |
| OpenRouterNot disclosed | OpenAI: GPT Terra LatestPrompt > 272K | $0.1/M cache read | input_cache_read: 0.0000001 / token (>272000) | |
| RunwareNot disclosed | Claude Haiku 4.5Per 1M | $0.1/M cache read | $1e-7 / cachedInputToken | |
| RunwareNot disclosed | Gemini 4 ArgonThrough Nov 30 · 2026 | $0.1/M cache read | $1e-7 / cachedInputToken · Through Nov 30, 2026 | |
| RunwareNot disclosed | GPT-6.1 Sol≤ 272K context | $0.1/M cache read | $1e-7 / cachedInputToken · ≤ 272K context | |
| RunwareNot disclosed | GPT-6.1 Terra> 272K context | $0.1/M cache read | $1e-7 / cachedInputToken · > 272K context | |
| OpenRouterOfficial | Qwen: Qwen3 Coder FlashPrompt > 128K | $0.104/M cache read | input_cache_read: 0.000000104 / token (>128000; endpoint alibaba) | |
| OpenRouterNot disclosed | Z.ai: GLM 5.2via DeepInfra · fp4 | $0.105/M cache read | input_cache_read: 0.000000105 / token (endpoint deepinfra/fp4) | |
| AibermNot disclosed | Kimi K3 | $0.106/M cache read | model_ratio 2.3 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| OpenRouterThird-party | Qwen: Qwen3 VL 235B A22B Instructvia DeepInfra · fp8 | $0.11/M cache read | input_cache_read: 0.00000011 / token (endpoint deepinfra/fp8) | |
| OpenRouterOfficial | Z.ai: GLM 4.5 | $0.11/M cache read | input_cache_read: 0.00000011 / token (endpoint z-ai/fp8) | |
| OpenRouterThird-party | Z.ai: GLM 4.5Vvia Novita · fp8 | $0.11/M cache read | input_cache_read: 0.00000011 / token (endpoint novita/fp8) | |
| AibermNot disclosed | GLM-5.3 | $0.11/M cache read | model_ratio 2.4 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | GLM-5 Turbo | $0.113/M cache read | model_ratio 1.224 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| OpenRouterOfficial | Baidu: ERNIE-Light-DT-1.0 | $0.113/M cache read | input_cache_read: 0.000000113 / token (endpoint baidu) | |
| AibermNot disclosed | Claude Fable 5.1 | $0.115/M cache read | model_ratio 5 × $2/M · group default ×0.23 · cache_ratio 0.05 | |
| AibermNot disclosed | MiniMax M2.1 | $0.115/M cache read | model_ratio 0.5 × $2/M · group default ×0.23 · cache_ratio 0.5 | |
| KIENot disclosed | grok-4-5 | $0.12/M cache read | 24 credits per million tokens ($0.12)Claims official or Fal price $0.3/M cache read | |
| OpenRouterThird-party | MoonshotAI: Kimi K2.7 Codevia CoreWeave · int4 | $0.12/M cache read | input_cache_read: 0.00000012 / token (endpoint coreweave/int4) | |
| OpenRouterThird-party | Qwen: Qwen3 VL 8B Instruct | $0.12/M cache read | input_cache_read: 0.00000012 / token (endpoint parasail/bf16) | |
| OpenRouterThird-party | Z.ai: GLM 5via StreamLake · fp8 | $0.12/M cache read | input_cache_read: 0.00000012 / token (endpoint streamlake/fp8) | |
| RunwareNot disclosed | MiniMax M3> 512K context | $0.12/M cache read | $1.2e-7 / cachedInputToken · > 512K context | |
| AibermNot disclosed | GPT-4o | $0.124/M cache read | model_ratio 0.9 × $2/M · group default ×0.23 · cache_ratio 0.3 | |
| OpenRouterOfficial | Anthropic: Claude Fable 5.1 (batch) | $0.125/M cache read | input_cache_read: 0.000000125 / token (endpoint anthropic) | |
| OpenRouterOfficial | Google: Gemini 2.5 Provia Google AI Studio | $0.125/M cache read | input_audio_cache: 0.000000125 / token (endpoint google-ai-studio) | |
| OpenRouterOfficial | Google: Gemini 2.5 Provia Google AI Studio | $0.125/M cache read | input_cache_read: 0.000000125 / token (endpoint google-ai-studio) | |
| OpenRouterOfficial | Google: Gemini 2.5 Pro Preview 06-05via Google AI Studio | $0.125/M cache read | input_audio_cache: 0.000000125 / token (endpoint google-ai-studio) | |
| OpenRouterOfficial | Google: Gemini 2.5 Pro Preview 06-05via Google AI Studio | $0.125/M cache read | input_cache_read: 0.000000125 / token (endpoint google-ai-studio) | |
| OpenRouterOfficial partner | OpenAI: GPT-5via Azure | $0.125/M cache read | input_cache_read: 0.000000125 / token (endpoint azure) | |
| OpenRouterOfficial partner | OpenAI: GPT-5.1-Codex-Max | $0.125/M cache read | input_cache_read: 0.000000125 / token (endpoint azure) | |
| OpenRouterOfficial | OpenAI: GPT-5.4 (batch) | $0.125/M cache read | input_cache_read: 0.000000125 / token (endpoint openai) | |
| AibermNot disclosed | MiniMax M2.5 | $0.128/M cache read | model_ratio 0.5555 × $2/M · group default ×0.23 · cache_ratio 0.5 | |
| AibermNot disclosed | GPT-5.6 | $0.13/M cache read | model_ratio 1.413 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| OpenRouterThird-party | DeepSeek: DeepSeek V3.1via DeepInfra · fp4 | $0.13/M cache read | input_cache_read: 0.00000013 / token (endpoint deepinfra/fp4) | |
| OpenRouterOfficial partner | OpenAI: GPT-5.1via Azure | $0.13/M cache read | input_cache_read: 0.00000013 / token (endpoint azure) | |
| OpenRouterOfficial partner | OpenAI: GPT-5.1-Codex | $0.13/M cache read | input_cache_read: 0.00000013 / token (endpoint azure) | |
| AibermNot disclosed | DeepSeek V4 Pro | $0.132/M cache read | model_ratio 1.43478 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| OpenRouterOfficial | OpenAI: o4 Mini (batch) | $0.138/M cache read | input_cache_read: 0.0000001375 / token (endpoint openai) | |
| AibermNot disclosed | GPT-5.1 | $0.138/M cache read | model_ratio 1 × $2/M · group default ×0.23 · cache_ratio 0.3 | |
| AibermNot disclosed | GPT-5.4 | $0.138/M cache read | model_ratio 1.5 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| AibermNot disclosed | GPT-5.5 xhigh | $0.138/M cache read | model_ratio 1.5 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| AibermNot disclosed | GPT-5.1 | $0.138/M cache read | model_ratio 1 × $2/M · group default ×0.23 · cache_ratio 0.3 | |
| AibermNot disclosed | GPT-5.4 | $0.138/M cache read | model_ratio 1.5 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| AibermNot disclosed | GPT-5.5 | $0.138/M cache read | model_ratio 1.5 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| KIENot disclosed | gpt-5.5 | $0.14/M cache read | 28 credits per million tokens ($0.14)Claims official or Fal price $0.5/M cache read | |
| KIENot disclosed | gpt-5.6-sol | $0.14/M cache read | 28 credits per million tokens ($0.14)Claims official or Fal price $0.5/M cache read | |
| AibermNot disclosed | Claude Sonnet 5 | $0.144/M cache read | model_ratio 3.1304 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | Claude Opus | $0.147/M cache read | model_ratio 3.19 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | Claude Fable 5.2 | $0.147/M cache read | model_ratio 12.8 × $2/M · group default ×0.23 · cache_ratio 0.025 | |
| OpenRouterOfficial | Anthropic: Claude Sonnet 4.5 (batch) | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint anthropic) | |
| OpenRouterOfficial | Anthropic: Claude Sonnet 4.6 (batch) | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint anthropic) | |
| OpenRouterOfficial | Cohere: Command A+ | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint cohere/fp8) | |
| OpenRouterOfficial | Google: Gemini 3.5 Flashvia Google · global | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint google-vertex/global) | |
| OpenRouterOfficial | Google: Gemini 3.5 Flash (batch) | $0.15/M cache read | input_audio_cache: 0.00000015 / token (endpoint google-vertex/global) | |
| OpenRouterOfficial | Meta: Muse Spark 1.1 | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint meta) | |
| OpenRouterOfficial | Meta: Muse Spark 1.2 | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint meta) | |
| OpenRouterThird-party | Meta: Muse Spark 1.3via Google · global | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint google-vertex/global) | |
| OpenRouterThird-party | Meta: Muse Spark Openvia Modal | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint modal) | |
| OpenRouterThird-party | MoonshotAI: Kimi K2 Thinking | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint novita/bf16) | |
| OpenRouterOfficial | Sakana: Sakana Namazu | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint sakana) | |
| OpenRouterThird-party | TheDrummer: Cydonia 24B V4.1 | $0.15/M cache read | input_cache_read: 0.00000015 / token (endpoint parasail/bf16) | |
| RunwareNot disclosed | Gemini 3.5 FlashPer 1M | $0.15/M cache read | $1.5e-7 / cachedInputToken | |
| RunwareNot disclosed | Gemini 3.6 FlashPer 1M | $0.15/M cache read | $1.5e-7 / cachedInputToken | |
| OpenRouterOfficial | Qwen: Qwen-PlusPrompt > 256K | $0.156/M cache read | input_cache_read: 0.000000156 / token (>256000; endpoint alibaba) | |
| OpenRouterOfficial | SpaceXAI: Grok 4.3 (batch) | $0.16/M cache read | input_cache_read: 0.00000016 / token (endpoint xai) | |
| RunwareNot disclosed | GLM-5.2Per 1M | $0.16/M cache read | $1.6e-7 / cachedInputToken | |
| AibermNot disclosed | Kimi K2.6 | $0.166/M cache read | model_ratio 1.8 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| AibermNot disclosed | Kimi K2.7 Code | $0.166/M cache read | model_ratio 1.8 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| AibermNot disclosed | GPT-4.1 | $0.166/M cache read | model_ratio 1.2 × $2/M · group default ×0.23 · cache_ratio 0.3 | |
| OpenRouterThird-party | Meta: Llama 4 Maverickvia Parasail · fp8 | $0.17/M cache read | input_cache_read: 0.00000017 / token (endpoint parasail/fp8) | |
| OpenRouterThird-party | Thinking Machines: Inkling | $0.17/M cache read | input_cache_read: 0.00000017 / token (endpoint together) | |
| OpenRouterOfficial partner | OpenAI: GPT-5.2via Azure | $0.175/M cache read | input_cache_read: 0.000000175 / token (endpoint azure) | |
| OpenRouterOfficial partner | OpenAI: GPT-5.2 Chat | $0.175/M cache read | input_cache_read: 0.000000175 / token (endpoint azure) | |
| OpenRouterOfficial partner | OpenAI: GPT-5.2-Codex | $0.175/M cache read | input_cache_read: 0.000000175 / token (endpoint azure) | |
| OpenRouterOfficial partner | OpenAI: GPT-5.3-Codexvia Azure | $0.175/M cache read | input_cache_read: 0.000000175 / token (endpoint azure) | |
| OpenRouterThird-party | Z.ai: GLM 5.2 (batch) | $0.178/M cache read | input_cache_read: 0.000000178 / token (endpoint morph/fp8) | |
| OpenRouterThird-party | Z.ai: GLM 5.3via Wafer · fp4 | $0.178/M cache read | input_cache_read: 0.000000178 / token (endpoint wafer/fp4) | |
| OpenRouterThird-party | Z.ai: GLM 5.3 (batch)via Morph · fp8 | $0.178/M cache read | input_cache_read: 0.000000178 / token (endpoint morph/fp8) | |
| OpenRouterOfficial | AionLabs: Aion-3.0-Mini | $0.18/M cache read | input_cache_read: 0.00000018 / token (endpoint aion-labs) | |
| OpenRouterOfficial | AionLabs: Aion 3.5 Mini | $0.18/M cache read | input_cache_read: 0.00000018 / token (endpoint aion-labs) | |
| OpenRouterNot disclosed | MoonshotAI: Kimi K3via Relace · fp4 | $0.18/M cache read | input_cache_read: 0.00000018 / token (endpoint relace/fp4) | |
| OpenRouterNot disclosed | MoonshotAI: Kimi Latest | $0.18/M cache read | input_cache_read: 0.00000018 / token | |
| AibermNot disclosed | Claude Fable 5.1 | $0.184/M cache read | model_ratio 16 × $2/M · group default ×0.23 · cache_ratio 0.025 | |
| AibermNot disclosed | Claude Fable 5 | $0.19/M cache read | model_ratio 4.13 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| OpenRouterOfficial | Qwen: Qwen3.7 PlusPrompt > 256Kvia Alibaba | $0.192/M cache read | input_cache_read: 0.000000192 / token (>256000; endpoint alibaba) |
