Compare API pricing before you build.
Auto-updated OpenRouter API list prices, context limits and output limits. This is a cost-and-capacity comparison, not a quality ranking or subscription price guide.
| Price/capacity index | Model | Index | Tokens/$ | Input $/1M | Output $/1M | Blended $/1M | Context | Max output | Price move | Index move | Source |
|---|---|---|---|---|---|---|---|---|---|---|---|
| #1 | OpenAI: GPT-6 Luna Pro (batch)Openai | 100 | 10.0M | $0.05 | $0.25 | $0.1 | 1.1M | 128k | NEW | NEW | OpenRouter |
| #2 | OpenAI: GPT-6 Luna (batch)Openai | 100 | 10.0M | $0.05 | $0.25 | $0.1 | 1.1M | 128k | NEW | NEW | OpenRouter |
| #3 | NVIDIA: Nemotron 3.5 LightningNvidia | 92.2 | 9.8M | $0.07 | $0.2 | $0.1025 | 262k | 236k | • | ↓ | OpenRouter |
| #4 | DeepSeek: DeepSeek V4.1 Flash (batch)Deepseek | 59.6 | 6.0M | $0.112 | $0.336 | $0.168 | 1.0M | 131k | • | ↓ | OpenRouter |
| #5 | DeepSeek: DeepSeek V4.1 FlashDeepseek | 55.3 | 5.3M | $0.04 | $0.64 | $0.19 | 1.0M | 944k | ↑ | ↑ | OpenRouter |
| #6 | OpenAI: GPT-6 Luna ProOpenai | 50 | 5.0M | $0.1 | $0.5 | $0.2 | 1.1M | 128k | NEW | NEW | OpenRouter |
| #7 | OpenAI: GPT-6 LunaOpenai | 50 | 5.0M | $0.1 | $0.5 | $0.2 | 1.1M | 128k | NEW | NEW | OpenRouter |
| #8 | Z.ai: GLM 5.3 FlashZ Ai | 44.7 | 4.2M | $0.15 | $0.5 | $0.2375 | 1.3M | 944k | • | ↓ | OpenRouter |
| #9 | NVIDIA: Nemotron 3.5 Content SafetyNvidia | 44.6 | 5.0M | $0.2 | $0.2 | $0.2 | 131k | 118k | • | ↓ | OpenRouter |
| #10 | Qwen: Qwen3.8 Omni FlashQwen | 43.4 | 4.3M | $0.15 | $0.47 | $0.23 | 1.0M | 131k | NEW | NEW | OpenRouter |
| #11 | Qwen: Qwen3.8 FlashQwen | 43.4 | 4.3M | $0.15 | $0.47 | $0.23 | 1.0M | 131k | • | ↓ | OpenRouter |
| #12 | DeepSeek: DeepSeek V4 Flash Vision ExpDeepseek | 31.8 | 3.0M | $0.22 | $0.66 | $0.33 | 1.0M | 944k | • | ↓ | OpenRouter |
| #13 | Z.ai: GLM 5.3 FlashXZ Ai | 17 | 1.7M | $0.37 | $1.25 | $0.59 | 1.0M | 131k | • | ↓ | OpenRouter |
| #14 | Cohere: Command A+Cohere | 14.9 | 1.7M | $0.3 | $1.5 | $0.6 | 192k | 64k | NEW | NEW | OpenRouter |
| #15 | Google: Gemini 3.8 Flash (batch)Google | 13.1 | 1.3M | $0.375 | $1.875 | $0.75 | 1.0M | 66k | • | ↓ | OpenRouter |
| #16 | Google: Gemini 3.7 Flash (batch)Google | 13.1 | 1.3M | $0.375 | $1.875 | $0.75 | 1.0M | 66k | • | ↓ | OpenRouter |
| #17 | Google: Gemini 3.8 FlashGoogle | 6.6 | 667k | $0.75 | $3.75 | $1.5 | 1.0M | 66k | • | ↓ | OpenRouter |
| #18 | Google: Gemini 3.7 FlashGoogle | 6.6 | 667k | $0.75 | $3.75 | $1.5 | 1.0M | 66k | • | ↓ | OpenRouter |
| #19 | SpaceXAI: Grok 4.7X Ai | 4.1 | 417k | $1.6 | $4.8 | $2.4 | 500k | 450k | • | ↓ | OpenRouter |
| #20 | Qwen: Qwen3.8 Max (0902)Qwen | 3.3 | 333k | $2 | $6 | $3 | 1.0M | 131k | • | ↓ | OpenRouter |
| #21 | SpaceXAI: Grok 4.6X Ai | 3.3 | 333k | $2 | $6 | $3 | 500k | 450k | • | ↓ | OpenRouter |
| #22 | SpaceXAI: Grok 4.5X Ai | 3.3 | 333k | $2 | $6 | $3 | 500k | 450k | • | ↓ | OpenRouter |
| #23 | Anthropic: Claude Opus 5.5 (batch)Anthropic | 2.5 | 250k | $2 | $10 | $4 | 1.0M | 128k | • | ↓ | OpenRouter |
| #24 | MoonshotAI: Kimi K3 (batch)Moonshotai | 2.1 | 219k | $2.28 | $11.4 | $4.56 | 1.0M | 16k | • | ↓ | OpenRouter |
| #25 | MoonshotAI: Kimi K3Moonshotai | 1.8 | 167k | $3 | $15 | $6 | 1.0M | 944k | • | ↓ | OpenRouter |
| #26 | Anthropic: Claude Opus 5.5Anthropic | 1.2 | 125k | $4 | $20 | $8 | 1.0M | 128k | • | ↓ | OpenRouter |
| #27 | Anthropic: Claude Fable 5.1 (batch)Anthropic | 1 | 100k | $5 | $25 | $10 | 1.0M | 128k | • | ↓ | OpenRouter |
| #28 | Anthropic: Claude Fable 5.1Anthropic | 0.5 | 50k | $10 | $50 | $20 | 1.0M | 128k | • | • | OpenRouter |
Dynamic flagship selection picks the newest paid text models from major providers on OpenRouter. Blended price uses a 3:1 input-to-output token mix in USD per million tokens. Tokens per dollar is 1,000,000 divided by that blended price. Value score (0-100) combines tokens-per-dollar with a capability factor from context window and max output tokens (log-scaled). It ranks list-price efficiency, not benchmark quality. The price/capacity index compares list-price efficiency from OpenRouter Models API; it does not measure answer quality, latency, safety or app-subscription value. Check the provider and route terms before production use.