Models

Open-weight models, serverless or dedicated

Call 5 models per token right now, or run any of 284 open-weight models on GPUs reserved for you. Pricing and specs are generated from the live catalog and Hugging Face, so they never drift.

Serverless

Pay per token, no commitment. 5 models from $0.20 / 1M input, up to 1M context.

Intelligence vs price
Chat models on sference
↖ Smarter for less
30
35
40
45
50
$0.25$1.00
↑ IndexBlended $ / 1M · log →

Artificial Analysis Intelligence Index v4.3 (0–100), measured across all providers by Artificial Analysis, not sference specifically. Blended price weights input and output 3:1. Source

Decisions models answer questions about an input (classify, route, score) and return probabilities instead of generated text. Billed on input tokens only.

Dedicated deployments

284 open-weight models we deploy on GPU capacity reserved for you, with a private endpoint, billed per GPU-hour. Models tagged serverless are also available per token above. How dedicated works

ModelParams
DeepSeek-V4.1-FlashserverlesstrendingvisionMoE
DeepSeek
763B
GLM-5.3-FlashserverlesstrendingvisionMoE
Z.ai
321B
GLM-5.3serverlesstrendingMoE
Z.ai
753B
Kimi-K3.1-PronewtrendingvisionMoE
Moonshot AI
2.78T
DeepSeek-V4-Flash-0731serverlesstrendingMoE
DeepSeek
304B
Qwen3.8-27Btrendingvision
Qwen
28B
Qwen3.8-Flash-NexttrendingvisionMoE
Qwen
180B
Kokoro-82Mtrending
hexgrad
—
pyannote—
gemma-4-31B-ittrendingvision
Google
31B
bge-m3trending
BAAI
—
OpenAI1.5B
OpenAI809M
gpt-oss-120btrendingMoE
OpenAI
117B
Qwen3.6-35B-A3BtrendingvisionMoE
Qwen
36B
Qwen3.5-9Btrendingvision
Qwen
9.7B
Kimi-K3trendingvisionMoE
Moonshot AI
2.78T
Google744M
OpenAI21B
Xiaomi MiMo311B
Qwen3-0.6Btrending
Qwen
752M
gemma-4-26B-A4B-ittrendingvisionMoE
Google
26B
DeepSeek684B
gemma-4-E4B-ittrendingvision
Google
8.0B
Unlimited-OCRvisionMoE
Baidu
3.3B
Qwen1.9B
DeepSeek-V4-FlashserverlessMoE
DeepSeek
291B
VoxCPM2trending
OpenBMB
2.3B
Muse-Glimmer-30Btrendingvision
Meta Muse
30B
gemma-3-4b-ittrendingvision
Google
4.3B
Qwen3.5-4Btrendingvision
Qwen
4.7B
Google303M
OpenBMB2.5B
Google12B
InklingvisionMoE
Thinking Machines
952B
MiniMax-M3.1-FlashtrendingvisionMoE
MiniMax
220B
DeepSeek1.6T
Meta Llama71B
GLM-5.2serverlessMoE
Z.ai
753B
Resemble AI—
gemma-4-E2B-ittrendingvision
Google
5.1B
Z.ai31B
Qwen31B
Qwen2.3B
StepFun604B
DeepSeek-OCRvisionMoE
DeepSeek
3.3B
NVIDIA627M
NVIDIA638M
OpenBMB9.4B
Qwen28B
Qwen873M
Google4.3B
Microsoft8.7B
Qwen8.8B
Qwen8.2B
Fish Audio4.6B
Qwen2.45T
MiniMax-M3visionMoE
MiniMax
427B
Google1000M
Qwen596M
Microsoft1.0B
Kimi-K2.6visionMoE
Moonshot AI
1.03T
BAAI568M
DeepSeek685B
Kimi-K2.5visionMoE
Moonshot AI
1.03T
DeepSeek3.4B
Qwen80B
Qwen403B
Microsoft2.7B
Qwen2.0B
Sesame1.6B
poolside118B
Moonshot AI1.03T
Qwen4.0B
PaddlePaddle959M
Qwen35B
Tencent2.0B
IBM258M
NVIDIA—
Qwen2.3B
Hugging Face3.1B
XingChen AGI31B
Cohere2.1B
Qwen7.6B
Qwen15B
Meta Llama109B
DeepSeek8.2B
Qwen4.4B
OpenBMB1.3B
Datalab5.3B
FunAudioLLM—
Google27B
Xiaomi MiMo1.02T
Moonshot AI1.03T
MiniMax229B
Qwen125B
Microsoft3.8B
Qwen11B
Hy3MoE
Tencent
299B
DeepSeek1.65T
PaddlePaddle959M
Meta Muse822B
Liquid AI24B
Google12B
LightOn1.0B
dots.ocrvision
dots.ocr
3.0B
Qwen28B
Tencent1.1B
Qwen81B
inclusionAI7.9B
Nari Labs1.6B
Qwen33B
MiniMax229B
OpenBMB35B
Tencent780B
NVIDIA561B
Qwen31B
Aleph Alpha78B
Qwen2.1B
Qwen480B
Resemble AI—
Qwen235B
Qwen4.0B
Qwen31B
DeepSeek685B
Google268M
Meta Llama279M
Liquid AI2.7B
Inkling-SmallvisionMoE
Thinking Machines
266B
Z.ai754B
Moondream9.3B
Qwen2.1B
ByteDance Seed8.3B
Moonshot AI1.03T
Liquid AI1.2B
intfloat560M
Z.ai110B
NVIDIA32B
DeepSeek685B
Kwaipilot35B
Z.ai754B
Qwen8.1B
Qwen36B
Microsoft596M
DeepSeek685B
NVIDIA8.9B
Mistral AI4.7B
Swiss AI Initiative8.1B
Google7.8B
IBM8.8B
Google27B
Qwen33B
poolside33B
Jina AI572M
Mistral AI119B
IndexTeam—
Z.ai358B
Reflection AI501B
Mistral AI—
Qwen4.0B
Mistral AI4.4B
Z.ai10B
Tencent30B
Jina AI3.8B
Z.ai358B
MiniMax229B
Microsoft5.6B
Qwen596M
Upstage250B
Meta Llama402B
Jina AI1.6B
NVIDIA32B
Moonshot AI49B
Tencent8.0B
Mistral AI8.9B
Moonshot AI16B
Qwen31B
DeepSeek685B
Qwen938M
Hugging Face2.2B
NVIDIA1.1B
inclusionAI127B
IBM48M
MiniMax229B
Cohere30B
LG AI Research237B
ZeroEntropy4.0B
NVIDIA33B
Mistral AI24B
StepFun199B
Kyutai—
Z.ai357B
inclusionAI1000B
Baidu4.7B
Baidu22B
Tencent80B
Qwen235B
Qwen2.1B
Qwen4.0B
Z.ai2.3B
Moondream9.3B
Qwen8.2B
Google4.3B
Canopy Labs3.8B
Z.ai10B
Nanonets3.8B
Mistral AI24B
Liquid AI8.5B
Qwen8.2B
Nomic475M
Alibaba NLP31B
NVIDIA124B
Qwen73B
Moonshot AI73B
Google27B
Google12B
Qwen33B
Xiaomi MiMo310B
Moonshot AI1.03T
GLM-4.5VvisionMoE
Z.ai
108B
Mistral AI24B
Xiaomi MiMo1.02T
Alibaba NLP305M
Qwen8.8B
Qwen236B
NVIDIA979M
Snowflake568M
DeepSeek685B
Upstage103B
MiniMax456B
inclusionAI125B
Qwen81B
ByteDance Seed36B
Meituan LongCat562B
Fara-7Bvision
Microsoft
8.3B
Qwen236B
Xiaomi MiMo311B
Mistral AI14B
NVIDIA32B
RolmOCRvision
Reducto
8.3B
Qwen31B
NVIDIA2.6B
Alibaba NLP150M
Baidu30B
OpenDataLab1.2B
Jina AI597M
Z.ai33B
StepFun201B
inclusionAI107B
Nomic7.1B
Alibaba NLP149M
Qwen235B
Meta Llama12B
StepFun10B
NVIDIA1.2B
Perplexity596M
GLM-4.6VvisionMoE
Z.ai
108B
Google5.4B
IBM32B
DeepSeek685B
OpenBMB8.2B
OpenAI22B
Ai28.3B
Mistral AI24B
Cohere111B
Liquid AI1.6B
Mistral AI125B
Intern-S1visionMoE
InternLM
241B
Tencent8.0B
BAAI2.5B
NVIDIA8.0B
Jina AI2.4B
Cohere8.6B
NVIDIA1.2B
IBM149M
ServiceNow15B
Meituan LongCat1.78T
ByteDance Seed7.5B
Mistral AI—
Swiss AI Initiative71B
NVIDIA4.3B
Mixedbread1.5B
EuroLLM23B
IBM8.2B

Specs and licenses from each model's Hugging Face page, synced 2026-11-18. Sorted by popularity: Hugging Face likes, trending, and downloads, plus OpenRouter usage.

Browse by maker · 64