Model catalog

Every model RelayX currently routes to, by modality. Pricing already includes the RelayX margin.

LLM

ModelUpstreamContextInput / 1M tokensOutput / 1M tokens
deepseek/deepseek-v4-flashdeepseek128,000$0.169$0.339
deepseek/deepseek-v4-prodeepseek128,000$0.526$1.052
anthropic/claude-4-7-sonnetaihubmix200,000$3.300$16.500
anthropic/claude-4-7-opusaihubmix200,000$5.500$27.500
openai/gpt-5aihubmix128,000$1.375$11.000
openai/gpt-5-miniaihubmix128,000$0.275$2.200
google/gemini-3-proaihubmix1,000,000$1.375$11.000
qwen/qwen3-maxaihubmix32,000$0.418$1.672
openai/gpt-4.1aihubmix1,000,000$2.200$8.800
openai/gpt-4.1-miniaihubmix1,000,000$0.440$1.760
openai/gpt-4.1-nanoaihubmix1,000,000$0.110$0.440
openai/gpt-5-nanoaihubmix400,000$0.055$0.440
openai/gpt-5-codexaihubmix400,000$1.375$11.000
openai/o4-miniaihubmix200,000$1.210$4.840
google/gemini-3-flashaihubmix1,000,000$0.330$2.750
google/gemini-3-flash-liteaihubmix1,000,000$0.275$1.650
xai/grok-4-fastaihubmix2,000,000$0.220$0.550
moonshot/kimi-k2aihubmix256,000$0.605$2.420
zai/glm-4-6aihubmix200,000$0.066$0.242
alibaba/qwen3-coder-plusaihubmix1,000,000$0.594$2.376
qwen/qwen3-vl-flashaihubmix254,000$0.023$0.231
qwen/qwen3-vl-plusaihubmix256,000$0.151$1.507
minimax/m2aihubmix196,000$0.110$0.110

Embeddings

ModelUpstreamContextDimensionsInput / 1M tokens
dashscope/text-embedding-v4dashscope8,1921024$0.077

TTS

ModelUpstreamDefault voicePer 1M chars
vox/index-tts-2vox专业解说$1.575
vox/voice-clone-v1vox$1.575
minimax/speech-2.6-hdminimaxprofessional-narrator$52.500

ASR

ModelUpstreamFeaturesPer minute
alibaba/paraformer-v2dashscopeword timestamps · diarization$0.0319