Model catalogue

Prices per 1M tokens. Routing suffixes: :fast :cheap :eu work on any family.

ModelVendorContextInputOutputp50 TTFT
lum-route:smartLumenal Router—$0.09$0.3696 ms
lumen-core-70bLumenal · proprietary128k$0.35$1.40121 ms
lumen-core-8bLumenal · proprietary32k$0.04$0.1648 ms
llama-4.2-90bMeta · open weights128k$0.12$0.48140 ms
qwen-turbo-nextAlibaba · open weights256k$0.06$0.24102 ms
deepseek-r2DeepSeek · open weights128k$0.27$1.10180 ms
gpt-5OpenAI400k$1.25$10.00210 ms
gpt-5-miniOpenAI400k$0.25$2.00130 ms
claude-sonnet-4.5Anthropic200k$3.00$15.00240 ms
gemini-3-flashGoogle1M$0.10$0.4088 ms
devstral-small-2Mistral · open weights128k$0.08$0.3095 ms
mistral-embed-v3Mistral8k$0.02—31 ms

38 models total · open-weight entries include checksummed weight mirrors (24 GB VRAM minimum in 4-bit). Embeddings, moderation and image endpoints listed below the fold.