Providers

48 labs. One API.

Frontier labs, open-weight foundations, specialist providers — aggregated behind api.aigateway.sh/v1. Same auth, same request shape, same billing line.

Google
embeddingimagetextvideo
→ Gemini 3.1 Pro→ Gemini 3 Flash→ Imagen 4
Modelsall 16 →
OpenAI
audio-sttaudio-ttsimagetext
→ GPT-5.4→ GPT-5.4 Mini→ Gpt-Oss-120b
Modelsall 13 →
Anthropic
text
→ Claude Opus 4.7→ Claude Sonnet 4.6→ Claude Haiku 4.5
Modelsall 6 →
Deepgram
audio-sttaudio-tts
→ Nova-3→ Aura-2-EN→ Aura-2-ES
Modelsall 5 →
Meta
imagemoderationtexttranslation
→ Llama-4-Scout-17b-16e-Instruct→ Llama-Guard-3-8b→ Llama-3.3-70b-Instruct-Fp8-Fast
Modelsall 21 →
Alibaba Qwen
embeddingreasoningtext
→ Qwen2.5-Coder-32b-Instruct→ Qwen3-Embedding-0.6b→ Qwen3-30b-A3b-Fp8
Modelsall 8 →
BAAI
embeddingrerankreranking
→ Bge-M3→ BGE-Reranker-Base→ Bge-Large-EN-V1.5
Modelsall 6 →
MiniMax
audio-ttstextvideo
→ M2.7→ Speech 2.8 HD→ Speech 2.8 Turbo
Modelsall 5 →
Black Forest Labs
image
→ Flux-2-Klein-9b→ Flux-2-Klein-4b→ Flux-2-DEV
Modelsall 4 →
Moonshot
text
→ Kimi K2.6→ Kimi-K2.5
Modelsall 2 →
xAI
text
→ Grok 4→ Grok 4 Fast
Modelsall 2 →
Hugging Face
classificationtext
→ Meta-Llama-3-8b-Instruct→ Mistral-7b-Instruct-V0.2→ Gemma-7b-IT
Modelsall 15 →
Mistral
audio-sttaudio-ttsmoderationtext
→ Voxtral TTS→ Mistral Small 4→ Voxtral Mini Transcribe Realtime
Modelsall 8 →
ByteDance
image
→ Seedream 4.0→ Seedream 4.5→ Seedream 5 Lite
Modelsall 4 →
Recraft
image
→ Recraft V4→ Recraft V4 Pro→ Recraft V4 Pro Vector
Modelsall 4 →
Microsoft
classificationimagetext
→ Phi-2→ Resnet-50→ ResNet-50
Modelsall 3 →
Alibaba
imagetext
→ Qwen 3 Max→ Qwen 3.5 397B A17B→ Wan 2.6 Image
Modelsall 3 →
Runway
imagevideo
→ RunwayML Gen-4.5→ Stable-Diffusion-V1-5-Img2img→ Stable-Diffusion-V1-5-Inpainting
Modelsall 3 →
openai
stttexttts
→ gpt-4o-transcribe→ o4-mini→ tts-1-hd
Modelsall 3 →
DeepSeek
reasoningtext
→ Deepseek-R1-Distill-Qwen-32b→ Deepseek-Math-7b-Instruct
Modelsall 2 →
Leonardo.AI
image
→ Lucid-Origin→ Phoenix-1.0
Modelsall 2 →
Perplexity
text
→ Sonar Reasoning Pro→ Sonar Deep Research
Modelsall 2 →
Inworld
audio-tts
→ TTS 1.5 Max→ TTS 1.5 Mini
Modelsall 2 →
PixVerse
video
→ Pixverse V5.6→ Pixverse V6
Modelsall 2 →
Vidu
video
→ Vidu Q3 Pro→ Vidu Q3 Turbo
Modelsall 2 →
AI4Bharat
texttranslation
→ Indictrans2-EN-Indic-1B→ IndicTrans2 EN→Indic 1B
Modelsall 2 →
Pipecat
audio-sttclassification
→ Smart-Turn-V2→ Smart Turn v2
Modelsall 2 →
minimax
musicvideo
→ music-2.6→ hailuo-2.3-fast
Modelsall 2 →
LLaVA
vision
→ Llava-1.5-7b-HF
Modelsall 1 →
Unum
vision
→ Uform-Gen2-Qwen-500m
Modelsall 1 →
PFN
embedding
→ Plamo-Embedding-1b
Modelsall 1 →
AI Singapore
text
→ Gemma-Sea-Lion-V4-27b-IT
Modelsall 1 →
Defog
text
→ Sqlcoder-7b-2
Modelsall 1 →
FBL
text
→ Una-Cybertron-7b-V2-Bf16
Modelsall 1 →
IBM
text
→ Granite-4.0-H-Micro
Modelsall 1 →
NVIDIA
text
→ Nemotron-3-120b-A12b
Modelsall 1 →
OpenChat
text
→ Openchat-3.5-0106
Modelsall 1 →
TheBloke
text
→ Discolm-German-7b-V1-Awq
Modelsall 1 →
TII
text
→ Falcon-7b-Instruct
Modelsall 1 →
TinyLlama
text
→ Tinyllama-1.1b-Chat-V1.0
Modelsall 1 →
Zhipu AI
text
→ Glm-4.7-Flash
Modelsall 1 →
Lykon
image
→ Dreamshaper-8-Lcm
Modelsall 1 →
Stability AI
image
→ Stable-Diffusion-XL-Base-1.0
Modelsall 1 →
MyShell
audio-tts
→ Melotts
Modelsall 1 →
Cartesia
audio-tts
→ Sonic 3
Modelsall 1 →
Cohere
audio-stt
→ Cohere Transcribe
Modelsall 1 →
AssemblyAI
audio-stt
→ Universal 3 Pro
Modelsall 1 →
facebook
object-detection
→ detr-resnet-50
Modelsall 1 →
Missing someone

Tell us and we'll add them.

New labs, new open-weight releases, regional providers — email the slug and we usually ship routing within 24 hours.

Request a provider