LLM Models
13 models
Data: OpenRouter and official sources · refreshed daily · as of 7/31/2026
Quality vs. Price ($/1M input)
Quality vs. Speed (tok/s)
OpenAIAnthropicGoogleMetaDeepSeekMistral AI
Score = MMLU×20% + GPQA×30% + HumanEval×25% + SWE-Bench×25%⭐ Top - average TECHAGENT user ratingtok/s - output tokens/sec via API
Don't see a model you know? Sign in to suggest it.