10,549 prices on 4,760 listings · last collected 58 min ago · 383 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
All 3,231Per second of output 5Per image 6Per million input tokens 1,112Per million output tokens 1,076Per million tokens 5Per million cached input tokens (read) 596Per million cache-write tokens 223Per thousand characters 2Per second of compute time 12Per run (default parameters) 186Other units (see the platform’s wording) 8
Showing 1–100 of 596
Per million cached input tokens (read)100
| Platform | Model · what it covers | USD | Platform’s own wording | Since |
|---|---|---|---|---|
| OpenRouterThird-party | DeepSeek: DeepSeek V4.1 Flash 1017 (batch) | $0.002/M cache readCheapest | input_cache_read: 0.000000002 / token | |
| OpenRouterOfficial | Meta: Muse Spark 1.2 Contributor | $0.002/M cache readCheapest | input_cache_read: 0.000000002 / token | |
| OpenRouterOfficial | Meta: Muse Spark 1.3 Contributor | $0.002/M cache readCheapest | input_cache_read: 0.000000002 / token | |
| AibermNot disclosed | DeepSeek V4.1 Flash | $0.0024/M cache read | model_ratio 0.26087 × $2/M · group default ×0.23 · cache_ratio 0.02 | |
| KIENot disclosed | DeepSeek V4.1 Flash | $0.0025/M cache read | 0.5 credits per million tokens ($0.0025) | |
| OpenRouterOfficial | Nex AGI: Nex-N2.5-Mini | $0.0025/M cache read | input_cache_read: 0.0000000025 / token | |
| OpenRouterOfficial | OpenAI: GPT-5 Nano (batch) | $0.0025/M cache read | input_cache_read: 0.0000000025 / token | |
| OpenRouterThird-party | Xiaomi: MiMo-V2.6-Flash | $0.0028/M cache read | input_cache_read: 0.0000000028 / token | |
| KIENot disclosed | gpt-6-luna | $0.003/M cache read | 0.6 credits per million tokens ($0.003)Claims official or Fal price $0.01/M cache read | |
| OpenRouterOfficial | inclusionAI: Ling 3.0 Tiny | $0.003/M cache read | input_cache_read: 0.000000003 / token | |
| OpenRouterOfficial | Particle AI: Particle-1 | $0.003/M cache read | input_cache_read: 0.000000003 / token | |
| OpenRouterThird-party | DeepSeek: DeepSeek V4.1 Flash (batch) | $0.0034/M cache read | input_cache_read: 0.00000000336 / token | |
| OpenRouterThird-party | Xiaomi: MiMo-V2.6-Pro | $0.0036/M cache read | input_cache_read: 0.0000000036 / token | |
| OpenRouterOfficial | Meta: Muse Ember | $0.004/M cache read | input_cache_read: 0.000000004 / token | |
| OpenRouterThird-party | inclusionAI: Ling 3.0 Flash | $0.0042/M cache read | input_cache_read: 0.0000000042 / token | |
| OpenRouterOfficial | Anthropic: Claude Haiku 5.5 (batch) | $0.005/M cache read | input_cache_read: 0.000000005 / token | |
| OpenRouterThird-party | DeepSeek: DeepSeek V4.1 Flash 1017 | $0.005/M cache read | input_cache_read: 0.000000005 / token | |
| OpenRouterOfficial partner | OpenAI: GPT-5 Nano | $0.005/M cache read | input_cache_read: 0.000000005 / token | |
| OpenRouterOfficial | OpenAI: GPT-6 Luna Pro (batch) | $0.005/M cache read | input_cache_read: 0.000000005 / token | |
| OpenRouterOfficial | OpenAI: GPT-6 Luna (batch) | $0.005/M cache read | input_cache_read: 0.000000005 / token | |
| OpenRouterOfficial | Upstage: Solar Mini 4 | $0.005/M cache read | input_cache_read: 0.000000005 / token | |
| OpenRouterNot disclosed | DeepSeek: DeepSeek Flash Latest | $0.005/M cache read | input_cache_read: 0.000000005 / token | |
| KIENot disclosed | gpt-5.6-luna | $0.0056/M cache read | 1.12 credits per million tokens ($0.0056)Claims official or Fal price $0.02/M cache read | |
| OpenRouterThird-party | Meituan: LongCat 2.0 | $0.006/M cache read | input_cache_read: 0.000000006 / token | |
| OpenRouterOfficial | Qwen: Qwen3.7 Flash | $0.006/M cache read | input_cache_read: 0.000000006 / token | |
| OpenRouterNot disclosed | Z.ai: GLM 5.2 | $0.006/M cache read | input_cache_read: 0.000000006 / token | |
| AibermNot disclosed | GPT-4o Mini | $0.0069/M cache read | model_ratio 0.05 × $2/M · group default ×0.23 · cache_ratio 0.3 | |
| AibermNot disclosed | Claude Haiku 5.5 | $0.0074/M cache read | model_ratio 0.16 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| OpenRouterThird-party | MiniMax: MiniMax M3.1 Flashoff-peak 1500–0100 UTC | $0.0075/M cache read | input_cache_read: 0.0000000075 / token (utc_start 1500, utc_end 100) | |
| OpenRouterThird-party | inclusionAI: Ling 3.1 Flash | $0.009/M cache read | input_cache_read: 0.000000009 / token | |
| OpenRouterOfficial partner | Anthropic: Claude Haiku 5.5 | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial | Google: Gemini 2.5 Flash Lite | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial | Mistral: Ministral 3 3B 2512 | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterThird-party | Mistral: Mistral Small 3.2 24B | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial | Mistral: Voxtral Small 24B 2507 | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial | OpenAI: GPT-5.4 Nano (batch) | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial | OpenAI: GPT-5.6 Luna Pro (batch) | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial | OpenAI: GPT-5.6 Luna (batch) | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial partner | OpenAI: GPT-6 Luna | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial partner | OpenAI: GPT-6 Luna Pro | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterOfficial | OpenAI: GPT-6 Luna Pro (batch)Prompt > 272K | $0.01/M cache read | input_cache_read: 0.00000001 / token (>272000) | |
| OpenRouterOfficial | OpenAI: GPT-6 Luna (batch)Prompt > 272K | $0.01/M cache read | input_cache_read: 0.00000001 / token (>272000) | |
| OpenRouterNot disclosed | Anthropic: Claude Haiku Latest | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| OpenRouterNot disclosed | OpenAI: GPT Luna Latest | $0.01/M cache read | input_cache_read: 0.00000001 / token | |
| RunwareNot disclosed | Claude Haiku 5.5≤ 100K context | $0.01/M cache read | $1e-8 / cachedInputToken · ≤ 100K context | |
| RunwareNot disclosed | DeepSeek-V4.1-FlashPer 1M | $0.01/M cache read | $1e-8 / cachedInputToken | |
| RunwareNot disclosed | DeepSeek-V4.1-Flash-1017Per 1M | $0.01/M cache read | $1e-8 / cachedInputToken | |
| RunwareNot disclosed | GPT-6 Luna≤ 272K context | $0.01/M cache read | $1e-8 / cachedInputToken · ≤ 272K context | |
| AibermNot disclosed | Claude Haiku 4.5 | $0.012/M cache read | model_ratio 0.25 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | Claude Haiku 4.5 20251001 | $0.012/M cache read | model_ratio 0.25 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | Claude Haiku 4.5 20251001 Thinking | $0.012/M cache read | model_ratio 0.25 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| OpenRouterThird-party | DeepSeek: DeepSeek V4 Flash 0731 (batch) | $0.012/M cache read | input_cache_read: 0.000000012 / token | |
| OpenRouterThird-party | inclusionAI: Ling 3.0 Flash VL | $0.012/M cache read | input_cache_read: 0.000000012 / token | |
| OpenRouterThird-party | NVIDIA: Nemotron 3 Nano 30B A3B | $0.012/M cache read | input_cache_read: 0.000000012 / token | |
| RunwareNot disclosed | Gemma 4 31BPer 1M | $0.012/M cache read | $1.2e-8 / cachedInputToken | |
| OpenRouterOfficial | Google: Gemini 3.1 Flash Lite (batch) | $0.013/M cache read | input_cache_read: 0.0000000125 / token | |
| OpenRouterOfficial | OpenAI: GPT-4.1 Nano (batch) | $0.013/M cache read | input_cache_read: 0.0000000125 / token | |
| OpenRouterOfficial | OpenAI: GPT-5 Mini (batch) | $0.013/M cache read | input_cache_read: 0.0000000125 / token | |
| OpenRouterThird-party | DeepSeek: DeepSeek V4 Flash Vision Exp | $0.014/M cache read | input_cache_read: 0.000000014 / token | |
| RunwareNot disclosed | DeepSeek-V4-Flash-0731Per 1M | $0.014/M cache read | $1.4e-8 / cachedInputToken | |
| KIENot disclosed | gpt-6-1-terra | $0.015/M cache read | 3 credits per million tokens ($0.015)Claims official or Fal price $0.05/M cache read | |
| OpenRouterOfficial | Google: Gemini 3.5 Flash Lite (batch) | $0.015/M cache read | input_audio_cache: 0.000000015 / token | |
| OpenRouterOfficial | Google: Gemini 3.5 Flash Lite (batch) | $0.015/M cache read | input_cache_read: 0.000000015 / token | |
| OpenRouterThird-party | IBM: Granite 4.2 8B | $0.015/M cache read | input_cache_read: 0.000000015 / token | |
| OpenRouterThird-party | MiniMax: MiniMax M3.1 Flash | $0.015/M cache read | input_cache_read: 0.000000015 / token (utc_start 100, utc_end 1500) | |
| OpenRouterOfficial | Mistral: Ministral 3 8B 2512 | $0.015/M cache read | input_cache_read: 0.000000015 / token | |
| OpenRouterOfficial | Mistral: Mistral Small 4 | $0.015/M cache read | input_cache_read: 0.000000015 / token | |
| OpenRouterOfficial | Nex AGI: Nex-N2.5-Pro | $0.015/M cache read | input_cache_read: 0.000000015 / token | |
| RunwareNot disclosed | MiniMax M3.1 FlashPer 1M | $0.015/M cache read | $1.5e-8 / cachedInputToken | |
| AibermNot disclosed | Claude Sonnet 5.5 | $0.015/M cache read | model_ratio 0.6522 × $2/M · group default ×0.23 · cache_ratio 0.05 | |
| OpenRouterOfficial | Qwen: Qwen3.8 Flash | $0.016/M cache read | input_cache_read: 0.000000016 / token | |
| OpenRouterOfficial | Qwen: Qwen3.8 Omni Flash | $0.016/M cache read | input_cache_read: 0.000000016 / token | |
| AibermNot disclosed | GPT-6.1 Terra | $0.016/M cache read | model_ratio 0.35325 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | GPT-4.1 Mini | $0.018/M cache read | model_ratio 0.13 × $2/M · group default ×0.23 · cache_ratio 0.3 | |
| OpenRouterThird-party | DeepSeek: DeepSeek V4 Flash 0423 | $0.018/M cache read | input_cache_read: 0.000000018 / token | |
| OpenRouterThird-party | DeepSeek: DeepSeek V4.1 Flash | $0.018/M cache read | input_cache_read: 0.000000018 / token | |
| AibermNot disclosed | MiMo V2 Flash | $0.019/M cache read | model_ratio 0.21 × $2/M · group default ×0.23 · cache_ratio 0.2 | |
| AibermNot disclosed | GPT-5.6 Luna | $0.019/M cache read | model_ratio 0.2826 × $2/M · group default ×0.23 · cache_ratio 0.15 | |
| OpenRouterOfficial | LiquidAI: LFM3-1.7B | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterOfficial | Mistral: Ministral 3 14B 2512 | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterOfficial | Mistral: Saba | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterThird-party | NVIDIA: Nemotron 3.5 Super VL | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterOfficial partner | OpenAI: GPT-5.4 Nano | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterOfficial partner | OpenAI: GPT-5.6 Luna | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterOfficial partner | OpenAI: GPT-5.6 Luna Pro | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterOfficial | OpenAI: GPT-5.6 Luna Pro (batch)Prompt > 272K | $0.02/M cache read | input_cache_read: 0.00000002 / token (>272000) | |
| OpenRouterOfficial | OpenAI: GPT-5.6 Luna (batch)Prompt > 272K | $0.02/M cache read | input_cache_read: 0.00000002 / token (>272000) | |
| OpenRouterOfficial partner | OpenAI: GPT-6 LunaPrompt > 272K | $0.02/M cache read | input_cache_read: 0.00000002 / token (>272000) | |
| OpenRouterOfficial partner | OpenAI: GPT-6 Luna ProPrompt > 272K | $0.02/M cache read | input_cache_read: 0.00000002 / token (>272000) | |
| OpenRouterOfficial | Qwen: Qwen3.7 FlashPrompt > 32K | $0.02/M cache read | input_cache_read: 0.00000002 / token (>32000) | |
| OpenRouterThird-party | Z.ai: GLM 5.3 Flash (batch) | $0.02/M cache read | input_cache_read: 0.00000002 / token | |
| OpenRouterNot disclosed | OpenAI: GPT Luna LatestPrompt > 272K | $0.02/M cache read | input_cache_read: 0.00000002 / token (>272000) | |
| RunwareNot disclosed | GPT-5.6 LunaPer 1M | $0.02/M cache read | $2e-8 / cachedInputToken | |
| RunwareNot disclosed | GPT-6 Luna> 272K context | $0.02/M cache read | $2e-8 / cachedInputToken · > 272K context | |
| OpenRouterThird-party | Tencent: Hy3off-peak 1600–0000 UTC | $0.021/M cache read | input_cache_read: 0.000000020625 / token (utc_start 1600, utc_end 0) | |
| AibermNot disclosed | DeepSeek V3.2 | $0.021/M cache read | model_ratio 0.45 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | DeepSeek V3.2 Exp | $0.021/M cache read | model_ratio 0.45 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | DeepSeek V3.2 Exp Thinking | $0.021/M cache read | model_ratio 0.45 × $2/M · group default ×0.23 · cache_ratio 0.1 | |
| AibermNot disclosed | GPT-5 Mini | $0.021/M cache read | model_ratio 0.15 × $2/M · group default ×0.23 · cache_ratio 0.3 | |
| OpenRouterOfficial | Tencent: Hy4 | $0.021/M cache read | input_cache_read: 0.000000021 / token |
