OpenRouter API Pricing Update: 19 Free Models, New GPT-6.1, and 12 Price Changes

Nineteen models are currently free on OpenRouter, spanning providers including Google, NVIDIA, Thinking Machines, Dots Studio, Recursive, XingChen, Poolside, Cohere, LiquidAI, and OpenRouter itself. NVIDIA leads with seven free models: Nemotron 3.5 Super VL, Nemotron 3.5 Lightning, Nemotron 3 Ultra, Nemotron 3 Nano Omni, and Nemotron 3.5 Content Safety, all with 1,000,000 or 256,000 token context windows. Google offers four free models: Lyria 3 Pro Preview, Lyria 3 Clip Preview, Gemma 4 26B A4B, and Gemma 4 31B, each with 1,048,576 or 262,144 token contexts. The Free Models Router from OpenRouter provides a 200,000 token context window at no cost.

Three new models were added to OpenRouter on 2026-10-28. OpenAI launched GPT-6.1 Terra Pro at $1.00 per million input tokens and $5.00 per million output tokens, with a 1,050,000 token context window. A batch version of GPT-6.1 Terra Pro is priced at $0.50 input and $2.50 output per million tokens. PrismML introduced Ternary Bonsai 2 9B, an open-weights model priced at $0.08 input and $0.13 output per million tokens with a 262,144 token context. One model was removed: NVIDIA Nemotron 3 Super (free).

Twelve models had price adjustments on 2026-10-28. Xiaomi MiMo-V2.5-Pro saw the largest increase, with input rising from $0.3045 to $0.48 and output from $0.609 to $1.80 per million tokens. DeepSeek Flash Latest cut input from $0.04 to $0.0182 but raised output from $0.50 to $1.80. DeepSeek V4 Flash 0731 dropped significantly, with input falling from $0.44 to $0.119 and output from $1.32 to $0.238. MoonshotAI Kimi Latest reduced output from $20.00 to $9.7533 while raising input from $0.159 to $0.19. Z.ai GLM 5.1 lowered input from $1.40 to $0.966 and output from $4.40 to $3.036. Qwen3.5-27B increased output from $2.16 to $2.60 while slightly reducing input from $0.27 to $0.26. DeepSeek V4 Flash 0423 raised input from $0.091 to $0.14 and output from $0.182 to $0.28. Google Gemma 3 4B increased both input from $0.04 to $0.05 and output from $0.05 to $0.10. Z.ai GLM Flash Latest reduced input from $0.0413 to $0.0379 and output from $0.0758 to $0.0771. DeepSeek Pro Latest held output steady at $8.00 while nudging input from $0.2374 to $0.241. Qwen3.6 35B A3B kept input at $0.10 and raised output from $0.90 to $0.95. MoonshotAI Kimi K3 maintained output at $20.00 while increasing input from $0.159 to $0.20.

The cheapest paid model by input price is OpenAI gpt-oss-20b at $0.018 per million input tokens and $0.09 per million output tokens, with a 131,072 token context window and open weights. DeepSeek Flash Latest follows at $0.0182 input and $1.80 output per million tokens with a 1,048,576 token context. inclusionAI Ling 3.0 Flash is priced at $0.021 input and $0.063 output per million tokens with a 262,144 token context and open weights. OpenAI GPT-5 Nano (batch) costs $0.025 input and $0.20 output per million tokens with a 400,000 token context. Meta Llama 3.2 1B Instruct is at $0.027 input and $0.201 output per million tokens with a 60,000 token context and open weights. Mistral Nemo rounds out the list at $0.029 input and $0.03 output per million tokens with a 131,072 token context and open weights.

Five models offer the largest context window of 2,000,000 tokens: Auto Router (Beta), Pareto Code Router, SpaceXAI Grok 4.20 Multi-Agent, SpaceXAI Grok 4.20, and Auto Router, all from OpenRouter or x-ai. These represent the maximum context capacity available on the platform as of 2026-10-28.