LLM Inference
70 providers — compare pricing, models, speed, and capabilities. Click any provider to see full model listings and token costs.
GPT-6 Astra, GPT-5.6 Sol/Terra/Luna — plus image, voice, and realtime APIs
Claude 5 — Fable 5.1, Opus 5, Sonnet 5, Haiku 4.5 with 1M context
Gemini 3.8 Flash — plus Veo 3.1 video, Lyria 3.5 music, and Gemini Omni
Mistral Medium 3.5, Mistral Small 4, Voxtral TTS, Mistral OCR 4 — European sovereign AI
Command A and Command R7B — enterprise RAG, North platform, and on-premise deployment
DeepSeek V4 Pro and V4.1-Flash — 1M context, open-weight, peak/off-peak pricing
Grok 4.6 — SpaceXAI with Imagine Image 2.0, Voice Transcribe 2.0, and 500K context
Qwen3 — 235B MoE open-weight frontier model with hybrid thinking and 1M context
Kimi K3 — 1T MoE open-weight model, strong agentic and coding benchmark scores
Hunyuan T1 reasoning model and Hunyuan Turbo — competitive frontier pricing
GLM-5.3 frontier model with 1M context — available on Amazon Bedrock
MiniMax M3 — 1M context frontier model available on Together AI and Amazon Bedrock
Stable Diffusion 3.5 Large and Stable Image Ultra — open-weight image generation
FLUX 1.1 Pro Ultra — the quality benchmark for commercial image generation
Kling v2.1 Master — cinematic AI video generation competing with Google Veo 3
Palmyra X5 — enterprise-grade LLM with 1M context, function calling, and on-premise deployment
MiMo-V2.5-Pro, MiMo-V2.5, MiMo-V2.5-Pro-UltraSpeed — reasoning-focused models from Xiaomi
Solar Pro 4, Solar Pro 3, Solar Pro 2 — Korean frontier models with strong multilingual performance
Nova-3 speech-to-text API — real-time and batch transcription at low latency
Universal-2 speech-to-text with speaker diarisation and audio intelligence
Jamba 1.6 — hybrid SSM-Transformer models with 256K context
ERNIE 5 — Baidu's frontier Chinese-English bilingual LLM API
Doubao Seed 2.1 — ByteDance's frontier reasoning LLM via Volcano Engine
Step-5 — Chinese frontier LLM with strong multimodal and reasoning capability
Dream Machine — high-quality video generation from text and image prompts
Gen-4 video generation API — cinematic quality text-to-video and image-to-video
MiniMax M3, Kimi K3, GLM-5.3, DeepSeek V4 Pro, gpt-oss-120B — plus GB200/B200 GPU clusters
LPU + LPX inference — high-throughput tokens per second, $350M Series A
H100/H200/B200/B300/GB300 on-demand — Serverless Training API, Forge 2026 conference
Agent API, Router API, Search API, Embeddings API — Sonar deprecated Sept 27, 2026
CS-4 wafer-scale — high-throughput inference on GPT OSS 120B and Qwen 3.8 27B
NVIDIA DGX Cloud Lepton — enterprise serverless inference on DGX infrastructure
GPU marketplace + inference API — Llama 4, DeepSeek V4 Pro, and H100/H200 on-demand
200+ models — DeepSeek V4.1 Flash, Kimi K3, Kling v3.0 video, agent sandbox
HGX B300, B200, H200, H100 — European AI cloud with GDPR-native data sovereignty
Ultra-cheap open-weight inference — Llama 4, DeepSeek V4 Pro, Qwen3 at rock-bottom prices
One API for every major LLM — 300+ models, route to the cheapest or fastest provider automatically
Run any open-source AI model via API — FLUX 1.1 Pro, Llama 4, video, audio, and custom models
Sub-second image generation — FLUX 1.1 Pro Ultra, video, audio, and B200/H200 GPU compute
Voyage 3.5 — strong MTEB retrieval quality for RAG and semantic search
Flash v2.5 sub-300ms voice — 32 languages, voice cloning, and the ElevenLabs Studio
Optimized inference microservices for NVIDIA GPUs — deploy any model on your own hardware
Serverless AI inference at the edge — run LLMs in 300+ Cloudflare data centers globally
Inference infrastructure for production ML — GLM, DeepSeek V4, Kimi K3, Nemotron, Muse
Acquired by NVIDIA — OctoAI inference is now available via NVIDIA NIM
Swiss AI inference with CHF pricing
European LLM inference with EUR pricing
Chinese LLM inference platform with CNY pricing
Serverless inference — Kimi K3, GLM-5.3, MiniMax M3, DeepSeek V4, Qwen3.5, Llama 4
European AI inference — Mistral Small and GLM-5.3-Flash, priced in EUR ex-VAT
Embeddings API gateway (Singapore) — Gemini Embedding models, USD pricing, no chat completions
OpenAI-compatible inference API — GLM-5.3, Kimi, DeepSeek, MiniMax
High-throughput inference on SambaNova RDU chips — gpt-oss-120b, GLM-5.3, Llama
Japan-based inference — zero data retention, JPY pricing, TEE-secured Kimi K3
EU sovereign AI gateway — Mistral, GLM-5.3, DeepSeek with per-model jurisdiction labels
EU inference on SambaNova chips — Gemma 4, gpt-oss-120b, Whisper, embeddings
EU inference — Dublin, Helsinki, Paris — GLM-5.3, Kimi K3, zero data retention
Inference platform + GPU pods — 785+ models, Amazon Nova, H200 capacity
API gateway / reseller — 33 text models at CloudGPU customer rates
API router / marketplace — 117 text models with per-route pricing
Managed inference API — 10 shared-endpoint models with per-token pricing
Muse brand — Muse Spark 1.3, Muse Glimmer, Muse Image, Muse Voice via Meta Model API
Open-weight model inference on TEE GPUs
Uncensored open-weight inference — RP, creative writing, and unrestricted model access
Serverless Inference API — 150K+ open models including Llama, Qwen, and FLUX
AWS-native LLM access — Nova, Claude, Llama, DeepSeek, MiniMax, Kimi and more
B300-SXM GPU instances — French sovereign cloud, Generative APIs with 1M free tokens
Microsoft's unified AI platform — GPT-6, Claude 5, Llama 4, and 1,800+ models via Azure
Managed open-weight inference from IONOS Cloud
Managed inference on Crusoe Cloud — DeepSeek V4, Gemma 4, GLM 5.1, Llama 3.3, Qwen3
Serverless inference on Gradient AI — Claude, GPT-5, DeepSeek V4, Kimi K2.6, Qwen 3.5, Llama 4
Japanese cloud AI inference — gpt-oss-120b and LLM-jp models, priced in JPY per 10k tokens
