OpenRouter API Prices: October 16, 2026

There are 18 free models available. Notable free models include Google: Lyria 3 Pro Preview and Google: Lyria 3 Clip Preview (both with 1,048,576 context tokens), Thinking Machines: Inkling Small (free) and Thinking Machines: Inkling (free) (both with 1,048,576 context tokens), and NVIDIA: Nemotron 3.5 Super VL (free), NVIDIA: Nemotron 3.5 Lightning (free), and NVIDIA: Nemotron 3 Ultra (free) (each with 1,000,000 context tokens).

Five new models were added. Meta: Muse Spark Open is priced at $1.25 per million input tokens and $4.25 per million output tokens with 1,048,576 context tokens and open weights. Moondream: Moondream 3 Preview and Moondream: Moondream 3.1 are each priced at $0.30 per million input tokens and $2.50 per million output tokens with 32,000 context tokens and open weights. NVIDIA: Nemotron 3.5 Super VL is priced at $0.15 per million input tokens and $0.50 per million output tokens with 1,048,576 context tokens and open weights. NVIDIA: Nemotron 3.5 Super VL (free) is available at $0.00 per million input and output tokens with 1,000,000 context tokens and open weights.

Two models were removed: Anthropic: Claude Sonnet 4 and Pareto 26.10 Preview.

Several models had price changes. Qwen: Qwen3.6 27B decreased input price from $0.32 to $0.25 per million tokens and output price from $3.20 to $2.00 per million tokens. DeepSeek: DeepSeek Flash Latest decreased input price from $0.0158 to $0.0103 per million tokens and output price from $0.891 to $0.72 per million tokens. MiniMax: MiniMax M2.5 decreased output price from $1.08 to $0.95 per million tokens while input price remained at $0.27 per million tokens. Z.ai: GLM 5 increased input price from $0.60 to $1.00 per million tokens and output price from $1.92 to $3.20 per million tokens. Google: Gemini 3.8 Flash increased input price from $0.75 to $0.825 per million tokens and output price from $3.75 to $4.125 per million tokens. Z.ai: GLM 5.3 increased input price from $0.10 to $0.40 per million tokens and output price from $4.80 to $8.00 per million tokens. Z.ai: GLM Flash Latest decreased input price from $0.036 to $0.0158 per million tokens while output price remained at $0.90 per million tokens. OpenAI: gpt-oss-120b increased input price from $0.037 to $0.15 per million tokens and output price from $0.17 to $0.60 per million tokens. DeepSeek: DeepSeek V4 Flash 0731 decreased input price from $0.016 to $0.0058 per million tokens while output price remained at $0.64 per million tokens. MoonshotAI: Kimi Latest decreased input price from $0.49 to $0.35 per million tokens and output price from $14.90 to $14.00 per million tokens. MoonshotAI: Kimi K3 decreased input price from $0.50 to $0.405 per million tokens while output price remained at $15.00 per million tokens. Z.ai: GLM 5.1 decreased input price from $1.33 to $0.966 per million tokens and output price from $4.18 to $3.036 per million tokens.

The cheapest paid models by input price are DeepSeek: DeepSeek V4 Flash Latest and DeepSeek: DeepSeek V4 Flash 0731, both at $0.0058 per million input tokens and $0.64 per million output tokens with 1,048,576 context tokens. The next cheapest are DeepSeek: DeepSeek Flash Latest and DeepSeek: DeepSeek V4.1 Flash at $0.0103 per million input tokens and $0.72 per million output tokens with 1,048,576 context tokens. Following these are Z.ai: GLM Flash Latest at $0.0158 per million input tokens and $0.90 per million output tokens with 1,048,576 context tokens, IBM: Granite 4.0 Micro at $0.017 per million input tokens and $0.112 per million output tokens with 131,000 context tokens, OpenAI: gpt-oss-20b at $0.018 per million input tokens and $0.09 per million output tokens with 131,072 context tokens, and inclusionAI: Ling 3.0 Flash at $0.021 per million input tokens and $0.063 per million output tokens with 262,144 context tokens.

The largest context windows are held by Auto Router (Beta), Pareto Code Router, SpaceXAI: Grok 4.20 Multi-Agent, SpaceXAI: Grok 4.20, and Auto Router, all offered by OpenRouter or xAI with 2,000,000 context tokens.