OpenRouter LLM API Pricing Update – 2026-11-03
OpenRouter’s LLM API catalog now tracks 451 models, with 17 remaining free, one new paid model added, and two models removed since yesterday.
Free Model Landscape
The platform continues to offer 17 free models, representing a subset of the 170 open‑weight models tracked. Free offerings span providers such as Google, NVIDIA, Thinking Machines, Dots Studio, Recursive, XingChen, Cohere, LiquidAI and the OpenRouter Free Models Router. Context windows for free models range from 65,536 tokens for LiquidAI: LFM2.5-2.6B up to 1,048,576 tokens for Google: Lyria 3 Pro Preview and Google: Lyria 3 Clip Preview.
New and Removed Models
A single new paid model entered the list: Z.ai: GLM 5.3 FlashX Legal, priced at $0.37 per million input tokens and $1.25 per million output tokens, with a 1,048,576‑token context window and no open‑weights flag. Two models were removed: Nous: Hermes 3 405B Instruct and Nous: Hermes 3 70B Instruct, both previously listed under the nousresearch provider.
Price Adjustments
Three models saw price changes since the previous day. Z.ai: GLM 5.3 input price rose from $0.0699 to $0.08 per million tokens and output price jumped from $1.4 to $3.6 per million tokens. Meta: Llama 3.2 3B Instruct input increased marginally from $0.05 to $0.0509 per million tokens, while output moved from $0.33 to $0.335 per million tokens. Google: Gemma 4 26B A4B input rose from $0.0675 to $0.0765 per million tokens and output from $0.225 to $0.255 per million tokens.
Cheapest Paid Options and Context Windows
The lowest input‑price paid models are Z.ai: GLM 5.2 at $0.006 per million input tokens (output $6.8), inclusionAI: Ling 3.0 Flash at $0.021 (output $0.063), and OpenAI: gpt-oss-20b (batch) at $0.024 (output $0.112). Other competitive options include Nex AGI: Nex-N2.5-Mini ($0.025 input, $0.1 output), OpenAI: GPT-5 Nano (batch) ($0.025 input, $0.2 output), Meta: Llama 3.2 1B Instruct ($0.027 input, $0.201 output), and Z.ai: GLM Flash Latest and Z.ai: GLM 5.3 Flash both at $0.0285 input and $0.0923 output. The largest context windows across the catalog are 2,000,000 tokens, offered by SpaceXAI: Grok 4.20 Multi‑Agent, SpaceXAI: Grok 4.20, and several OpenRouter routers (Auto Router, Auto Router (Beta), Pareto Code Router).
Key takeaways
- Free model count stays at 17 while the total catalog holds 451 models.
- One new paid model was added and two models were removed since the prior day.
- Three models experienced price adjustments, with Z.ai: GLM 5.3 seeing the largest increase in output cost.
- The cheapest input‑price options start at $0.006 per million tokens, and the maximum context window remains 2,000,000 tokens.