💲 LLM API Pricing
Chat per-token pricing plus image, video, speech, transcription and embedding rates — refreshed daily · updated 12h. Data: LiteLLM + OpenRouter, plus a small curated set (Kling, Hailuo, ElevenLabs, Cartesia) from official pricing pages.
All providersbedrock_converse 599bedrock_mantle 419fireworks_ai 328bedrock 275azure 211deepinfra 140novita 130gmi 122azure_ai 116openai 104vercel_ai_gateway 93empiriolabs 92kyma 83replicate 83
Cost calculator — what would your workload cost on each model?
| Model | Elo | Input $/M | Output $/M | Cache read | Cache write | Batch in/out | Context | History |
|---|---|---|---|---|---|---|---|---|
| FW-GPT-OSS-120Bazure_ai | — | $0.17 | $0.66 | $0.0820 | — | — | 131K | |
| mistral-nemo💬azure_ai | — | $0.15 | $0.15 | — | — | — | 131K | |
| FW-DeepSeek-V4-Flashazure_ai | — | $0.15 | $0.31 | $0.0300 | — | — | 1.0M | |
| gpt-oss-120b💬azure_ai | 1365 | $0.15 | $0.60 | — | — | — | 131K | |
| model_routerazure_ai | — | $0.14 | free | — | — | — | 200K | |
| model-routerazure_ai | — | $0.14 | free | — | — | — | 200K | |
| Phi-4-reasoningazure_ai | — | $0.13 | $0.50 | — | — | — | 33K | |
| Phi-4💬azure_ai | 1217 | $0.13 | $0.50 | — | — | — | 16K | |
| mistral-small-2503azure_ai | — | $0.10 | $0.30 | — | — | — | 128K | |
| claude-haiku-5-5💬azure_ai | — | $0.10 | $0.50 | $0.0100 | $0.13 | — | 1.0M | |
| gpt-6-luna💬azure_ai | — | $0.10 | $0.50 | $0.0100 | $0.13 | — | 922K | |
| Phi-4-multimodal-instructazure_ai | — | $0.0800 | $0.32 | — | — | — | 131K | |
| Phi-4-mini-instruct💬azure_ai | — | $0.0750 | $0.30 | — | — | — | 131K | |
| Phi-4-mini-reasoningazure_ai | — | $0.0750 | $0.30 | — | — | — | 131K | |
| FW-Nemotron-Lightning-3.5-30B-A3Bazure_ai | — | $0.0600 | $0.22 | $0.0100 | — | — | 262K | |
| ministral-3bazure_ai | — | $0.0400 | $0.0400 | — | — | — | 128K |
Recent price changes
- ▼ 53.3%Qwen: Qwen3 Coder 30B A3B Instructqwen · output $0.60 → $0.287h
- ▼ 53.3%Qwen: Qwen3 Coder 30B A3B Instructqwen · input $0.15 → $0.07007h
- ▼ 40.0%Qwen: Qwen3.5-27Bqwen · output $2.60 → $1.567h
- ▼ 25.0%Qwen: Qwen3.5-27Bqwen · input $0.26 → $0.207h
- ▼ 5.6%OpenAI: gpt-oss-20bopenai · output $0.0900 → $0.08507h
- ▼ 5.6%OpenAI: gpt-oss-20bopenai · input $0.0180 → $0.01707h
- ▼ 6.0%NVIDIA: Nemotron 3.5 Super VLnvidia · output $0.50 → $0.477h
- ▲ 114.3%MiniMax: MiniMax M2.7minimax · output $0.84 → $1.807h
- ▲ 185.7%MiniMax: MiniMax M2.7minimax · input $0.21 → $0.607h
- ▼ 20.8%MiniMax: MiniMax M2.5minimax · output $1.20 → $0.957h
- ▼ 8.5%MiniMax: MiniMax M2.5minimax · input $0.29 → $0.277h
- ▲ 33.3%inclusionAI: Ling 3.1 Flashinclusionai · output $0.14 → $0.187h
- ▲ 33.3%inclusionAI: Ling 3.1 Flashinclusionai · input $0.0450 → $0.06007h
- ▼ 11.8%Google: Gemma 4 26B A4B google · output $0.26 → $0.237h
- ▼ 11.8%Google: Gemma 4 26B A4B google · input $0.0765 → $0.06757h
- ▲ 181.3%Google: Gemma 3 27Bgoogle · output $0.16 → $0.457h
- ▼ 51.0%DeepSeek: DeepSeek V4 Flash Vision Expdeepseek · output $1.32 → $0.657h
- ▼ 51.0%DeepSeek: DeepSeek V4 Flash Vision Expdeepseek · input $0.44 → $0.227h
- ▼ 26.7%DeepSeek: DeepSeek V4 Flash Latestdeepseek · output $0.18 → $0.137h
- ▼ 26.7%DeepSeek: DeepSeek V4 Flash Latestdeepseek · input $0.0600 → $0.04407h
Price cuts are green — they make AI cheaper for everyone.