LLM & Generative AI
Text, video & audio model pricing & performance.
14 providers measured 5 data categories ranked
Top 3 in LLM & Generative AI
View full rankings →
step-5-preview 
hy4-preview GPT-4o
OpenAI's most popular chat model. Every listing here is a gpt-4o variant sold through the OpenRouter hub.
100% available p50 latency 128 ms
cheapestfreefastestreliableoverall
Claude Sonnet
Anthropic's balanced everyday model — fast and capable, with a batch (50% off) option on the hub.
100% available p50 latency 128 ms
cheapestfreefastestreliableoverall
Gemini
Google's family of multimodal models, from the small Flash-Lite up to the big Pro tiers.
100% available p50 latency 128 ms
cheapestfreefastestreliableoverall
DeepSeek
Known for the lowest prices around. The V3 / V4 / R1 lines on the hub, plus DeepSeek's own endpoint.
50% available p50 latency 279 ms
cheapestfreefastestreliableoverall
Qwen
Alibaba's open-weight family — cheap Flash sizes for everyday use up to the big 3.x generation.
100% available p50 latency 128 ms
cheapestfreefastestreliableoverall