OpenRouter API Pricing Report, 2026-11-07

Fifteen models are free on 2026-11-07. Crouton Alpha, Google: Lyria 3 Pro Preview, and Google: Lyria 3 Clip Preview each offer 1048576 context tokens. NVIDIA: Nemotron 3.5 Super VL (free), NVIDIA: Nemotron 3.5 Lightning (free), and NVIDIA: Nemotron 3 Ultra (free) each offer 1000000 context tokens. Dots Studio: Dots3-Note Preview (free) offers 512000 context tokens. Recursive: Flint (free), XingChen: Xing4.0 29B A4B (free), Google: Gemma 4 26B A4B (free), and Google: Gemma 4 31B (free) each offer 262144 context tokens. Cohere: North Mini Code (free) and NVIDIA: Nemotron 3 Nano Omni (free) offer 256000 context tokens. Free Models Router offers 200000 context tokens. LiquidAI: LFM2.5-2.6B (free) offers 65536 context tokens.

Five new batch models appeared: DeepSeek: DeepSeek V4 Flash 0731 (batch) at $0.048 input and $0.144 output per million tokens with 1048576 context tokens; Meta: Muse Glimmer 30B (batch) at $0.24 input and $0.96 output per million tokens with 131072 context tokens; MiniMax: MiniMax M3 (batch) at $0.224 input and $0.88 output per million tokens with 524288 context tokens; Qwen: Qwen3.8 27B (batch) at $0.16 input and $2.0 output per million tokens with 262144 context tokens; Thinking Machines: Inkling Small (batch) at $0.36 input and $0.96 output per million tokens with 524288 context tokens. NVIDIA: Nemotron 3.5 Content Safety (free) was removed. No models became free or stopped being free since 2026-11-06.

Twelve models changed prices. DeepSeek: DeepSeek V4.1 Flash input rose from $0.045 to $0.051 per million tokens and output rose from $0.5 to $0.567 per million tokens. DeepSeek: DeepSeek Pro Latest input rose from $0.0927 to $0.0999 per million tokens while output stayed at $8.0 per million tokens. NVIDIA: Nemotron 3 Super input fell from $0.3 to $0.075 per million tokens and output fell from $2.4 to $0.4 per million tokens. DeepSeek: DeepSeek V4 Pro 0423 input fell from $1.6 to $0.6892 per million tokens and output fell from $3.2 to $1.3784 per million tokens. DeepSeek: DeepSeek V4 Flash Latest input fell from $0.059 to $0.0434 per million tokens while output stayed at $0.3 per million tokens. DeepSeek: DeepSeek V4 Pro 0813 input rose from $0.103 to $0.1165 per million tokens and output rose from $4.11 to $4.66 per million tokens. MoonshotAI: Kimi Latest output rose from $9.3388 to $10.1489 per million tokens while input stayed at $0.1824 per million tokens. DeepSeek: DeepSeek V4 Flash 0731 input fell from $0.059 to $0.0434 per million tokens while output stayed at $0.3 per million tokens. Google: Gemma 4 26B A4B input fell from $0.0765 to $0.0675 per million tokens and output fell from $0.255 to $0.225 per million tokens. OpenAI: gpt-oss-120b input fell from $0.023 to $0.016 per million tokens and output fell from $0.084 to $0.061 per million tokens. Z.ai: GLM Flash Latest input fell from $0.0285 to $0.0284 per million tokens while output stayed at $0.0923 per million tokens. DeepSeek: DeepSeek V4 Flash Vision Exp input rose from $0.2156 to $0.44 per million tokens and output rose from $0.6468 to $1.32 per million tokens.

The cheapest paid models by input price are DeepSeek: DeepSeek Flash Latest at $0.006 input and $0.5 output per million tokens with 1048576 context tokens, and Z.ai: GLM 5.2 at $0.006 input and $6.8 output per million tokens with 1048576 context tokens. inclusionAI: Ling 3.0 Tiny costs $0.015 input and $0.045 output per million tokens with 262144 context tokens. OpenAI: gpt-oss-120b costs $0.016 input and $0.061 output per million tokens with 131072 context tokens. inclusionAI: Ling 3.0 Flash costs $0.021 input and $0.063 output per million tokens with 262144 context tokens. OpenAI: gpt-oss-20b (batch) costs $0.024 input and $0.112 output per million tokens with 131072 context tokens. Nex AGI: Nex-N2.5-Mini costs $0.025 input and $0.1 output per million tokens with 262144 context tokens. OpenAI: GPT-5 Nano (batch) costs $0.025 input and $0.2 output per million tokens with 400000 context tokens.

Five models offer 2000000 context tokens: Auto Router, Auto Code, Auto Router (Beta), and Pareto Code Router, all from openrouter, plus SpaceXAI: Grok 4.20 Multi-Agent from x-ai.