Providers
60 providers with at least one ranked model. Ordered by their best configuration on the BenchLeader Index.
- 1
Anthropic72.7
- Best model
- Claude Opus 5.5
- Ranked
- 23
- From
- $0.200/M
Safety-focused lab behind the Claude family (Haiku, Sonnet, Opus and the Claude 5 tier). Strong in coding and agentic work; sells API access, Claude apps and Claude Code.
- 2
Google71.0
- Best model
- Gemini 4 Argon
- Ranked
- 41 · 16 open
- From
- $0.040/M
Google DeepMind builds the Gemini models (Pro, Flash, Flash-Lite, Deep Think) and the open Gemma family; distributed through Gemini apps, Vertex AI and AI Studio.
- 3
Meta70.7
- Best model
- Muse Wonder Full
- Ranked
- 27 · 19 open
- From
- $0.048/M
Ships the open-weights Llama family and Meta AI products; announced the Superintelligence Labs reorganisation in 2025 and very large data-centre investments.
- 4
OpenAI70.6
- Best model
- GPT-6 Astra
- Ranked
- 45 · 3 open
- From
- $0.115/M
Maker of the GPT and o-series models and ChatGPT. Sells API access, ChatGPT subscriptions and enterprise deals; deeply tied to Microsoft Azure.
- 5
Moonshot AI64.9
- Best model
- Kimi K3
- Ranked
- 8 · 9 open
- From
- $1.00/M
Maker of the Kimi assistant and the open-weights Kimi K2 / K3 mixture-of-experts models with strong agentic scores.
- 6
SpaceXAI64.2- Best model
- Grok 4.6
- Ranked
- 16 · 1 open
- From
- $0.275/M
- 7
Xiaomi64.2
- Best model
- MiMo-V2.6-Pro
- Ranked
- 7 · 5 open
- From
- $0.150/M
MiMo models from Xiaomi's AI lab, aimed at on-device and assistant use.
- 8
Zhipu AI63.6
- Best model
- GLM 5.3
- Ranked
- 14 · 15 open
- From
- $0.153/M
Tsinghua spin-out behind the GLM series, released open-weights (GLM-4.5, 4.6, 5) with a focus on agentic coding.
- 9
Alibaba63.3
- Best model
- Qwen3.8 Max
- Ranked
- 54 · 50 open
- From
- $0.060/M
Alibaba Cloud's Qwen team ships the broadest open-weights line-up (Qwen3, Qwen3.5, coder and VL variants) plus hosted Qwen Max models.
- 10
DeepSeek63.1
- Best model
- DeepSeek V4.1 Flash 1017
- Ranked
- 20 · 22 open
- From
- $0.068/M
Open-weights lab known for cost-efficient training (V3, R1, V4) and aggressive API pricing.
- 11
MiniMax61.9
- Best model
- MiniMax-M3.1-Flash
- Ranked
- 7 · 9 open
- From
- $0.262/M
Consumer-app-driven lab (Talkie, Hailuo video) that also ships the open-weights MiniMax M-series text models.
- 12
StepFun60.6
- Best model
- Step 5 Preview
- Ranked
- 5 · 4 open
- From
- $0.150/M
Step series multimodal and reasoning models.
- 13
Sapiens AI59.1
- Best model
- Agnes 2.5 Pro Beta
- Ranked
- 1
- From
- $0.150/M
- 14
ByteDance58.7
- Best model
- Dola Seed 2.0 Pro
- Ranked
- 3 · 1 open
- From
- $0.300/M
ByteDance Seed builds the Doubao/Seed models powering the Doubao assistant, one of China's most used chatbots.
- 15
Motif Technologies57.8- Best model
- Motif 3
- Ranked
- 2 · 1 open
- From
- –
- 16
Mistral AI56.8
- Best model
- Mistral Large 4
- Ranked
- 31 · 32 open
- From
- $0.058/M
Europe's leading lab, mixing open-weights (Mistral Small, Devstral) and commercial models (Medium, Large, Magistral); strong enterprise and sovereign-cloud positioning.
- 17
MBZUAI56.3- Best model
- K2 Horizon 375B A23B
- Ranked
- 6 · 6 open
- From
- –
- 18
Meituan56.2
- Best model
- Longcat Flash
- Ranked
- 3 · 2 open
- From
- $1.30/M
- 19
Multiverse Computing55.8- Best model
- Quasar 438B
- Ranked
- 2 · 1 open
- From
- $0.900/M
- 20
NVIDIA55.6
- Best model
- Nemotron 3 Ultra 550B A55B
- Ranked
- 15 · 25 open
- From
- $0.070/M
The GPU maker also publishes open Nemotron models fine-tuned from Llama and its own architectures, mostly as showcases for its inference stack.
- 21
Ant Group55.6
- Best model
- Ling 3.0 Flash VL
- Ranked
- 9 · 10 open
- From
- $0.111/M
- 22Thinking Machines54.9
- Best model
- Inkling Small
- Ranked
- 1 · 1 open
- From
- $0.525/M
- 23
Nex AGI54.7- Best model
- Nex-N2-Pro
- Ranked
- 1 · 1 open
- From
- $1.00/M
- 24
Reflection54.6- Best model
- Beam 1.0
- Ranked
- 1 · 2 open
- From
- $1.05/M
- 25Apodex54.4
- Best model
- Apodex 1.1
- Ranked
- 1
- From
- $0.975/M
- 26
Upstage53.6
- Best model
- Solar Open2 250B
- Ranked
- 4 · 4 open
- From
- $0.262/M
- 27
SK Telecom52.6- Best model
- A.X-K2
- Ranked
- 1 · 1 open
- From
- –
- 28
KwaiKAT52.4
- Best model
- KAT-Coder-Pro V2
- Ranked
- 1
- From
- –
- 29China Mobile52.0
- Best model
- JT-4.1 Flash 236B A21B
- Ranked
- 3
- From
- –
- 30AI9Stars52.0
- Best model
- G9v3-39A5B
- Ranked
- 1 · 2 open
- From
- –
- 31
Cohere50.6
- Best model
- Command A+
- Ranked
- 6 · 17 open
- From
- $0.600/M
Enterprise-focused lab behind Command A and the Aya multilingual models, with a private-deployment emphasis.
- 32
Amazon50.5
- Best model
- Nova 2.0 Pro Preview
- Ranked
- 8
- From
- $0.065/M
AWS's Nova family sits alongside Bedrock's third-party catalogue; Amazon is also Anthropic's largest investor.
- 33
LG AI Research49.7
- Best model
- K-EXAONE
- Ranked
- 5 · 5 open
- From
- –
- 34
OpenBMB49.3- Best model
- MiniCPM5-2B
- Ranked
- 3 · 3 open
- From
- –
- 35
IBM48.8
- Best model
- Granite 4.2 30B
- Ranked
- 14 · 14 open
- From
- $0.053/M
Granite open models aimed at enterprise deployment through watsonx.
- 36
ServiceNow48.1- Best model
- Apriel-v1.6-15B-Thinker
- Ranked
- 1 · 1 open
- From
- –
- 37
Tencent47.6
- Best model
- Hunyuan Large Vision
- Ranked
- 1 · 4 open
- From
- –
Hunyuan models (including the T1 reasoning line) power Tencent's Yuanbao assistant and WeChat integrations.
- 38
Baidu47.4
- Best model
- ERNIE 5.0
- Ranked
- 2 · 1 open
- From
- –
ERNIE models (4.5 open-weights, 5 series) behind the ERNIE Bot assistant and Baidu Cloud.
- 39
Inception47.0
- Best model
- Mercury 2
- Ranked
- 2
- From
- $0.375/M
Diffusion-LLM pioneer; Mercury models generate tokens in parallel for very high output speed.
- 40
Arcee AI47.0
- Best model
- Trinity Large
- Ranked
- 1 · 1 open
- From
- $0.388/M
- 41
01.AI46.9
- Best model
- Yi Lightning
- Ranked
- 3
- From
- –
- 42
Kuaishou46.8- Best model
- Kat Coder Pro v1
- Ranked
- 1
- From
- –
- 43
Deep Cogito46.6
- Best model
- Cogito v2.1
- Ranked
- 1 · 1 open
- From
- –
- 44
Prime Intellect46.5- Best model
- Intellect 3
- Ranked
- 1 · 1 open
- From
- –
- 45
Reka44.6- Best model
- Reka Core
- Ranked
- 3 · 1 open
- From
- $0.350/M
- 46
Ai244.1
- Best model
- Olmo 3.1 32B
- Ranked
- 5 · 5 open
- From
- –
- 47
Liquid AI43.6
- Best model
- LFM2.5-8B-A1B
- Ranked
- 7 · 8 open
- From
- $0.053/M
- 48
Nous Research43.4
- Best model
- Hermes 4 - Llama-3.1 405B
- Ranked
- 2 · 5 open
- From
- $1.50/M
- 49
Trillion Labs43.4- Best model
- Tri-21B-Think
- Ranked
- 1 · 1 open
- From
- –
- 50
Naver42.9- Best model
- HyperCLOVA X SEED Think (32B)
- Ranked
- 1 · 1 open
- From
- –
- 51
TII42.6
- Best model
- Falcon-H1R-7B
- Ranked
- 1 · 2 open
- From
- –
- 52
KT42.5- Best model
- Mi Dm K 2.5 Pro Dec28
- Ranked
- 1
- From
- –
- 53
AI21 Labs42.3
- Best model
- Jamba 1.5 Large
- Ranked
- 7 · 7 open
- From
- $0.250/M
- 54Celeris41.4
- Best model
- Celeris-1
- Ranked
- 1
- From
- $0.325/M
- 55Nanbeige40.9
- Best model
- Nanbeige4.1-3B
- Ranked
- 1 · 1 open
- From
- –
- 56
Sarvam40.2- Best model
- Sarvam 105B
- Ranked
- 2 · 2 open
- From
- $0.073/M
- 57
Microsoft39.0
- Best model
- Phi-4
- Ranked
- 4 · 9 open
- From
- $0.135/M
Publishes the small open Phi models and MAI in-house models alongside its OpenAI partnership and Azure AI Foundry distribution.
- 58
OpenChat39.0
- Best model
- Openchat 3.5
- Ranked
- 1 · 1 open
- From
- –
- 59
Swiss AI38.6- Best model
- Apertus 70B Instruct
- Ranked
- 2 · 2 open
- From
- $0.125/M
- 60
Poolside37.4
- Best model
- Laguna M.1
- Ranked
- 2 · 3 open
- From
- –
Other labs
30 more labs with benchmark results but no model that has enough of them to rank yet.
LMSYS- Nexusflow
Databricks
UW NLPShanghai AI Lab
MosaicML
- Open Assistant
WriterRWKV
Hugging Face
Snowflake
Stability AI
Together AI
Cognitive Computations
- Marin
- LM-Provers
Cursor
- katanemo
- SpeakLeash & ACK Cyfronet AGH
Bittensor- UIUC + Oumi
- Dots Studio
- Tsinghua University
- MadeAgents
All Hands AI
Perplexity
- Phronetic AI
Cognition
Huawei
Salesforce