Catalog
Labs
Organisations that publish frontier model evaluations. Each lab page shows progress by benchmark, without combining unrelated benchmark metrics.
01
OpenAI
Model lab for OpenAI public benchmark rows.
24models539results
02Anthropic
Model lab for Claude public benchmark rows.
18models484results
03Google DeepMind
Model lab for Gemini public benchmark rows.
14models302results
04Meta
Model lab for Llama and Meta public benchmark rows.
8models108results
05DeepSeek
Model lab for DeepSeek public benchmark rows.
10models169results
06xAI
Model provider for Grok-family public benchmark rows.
8models83results
07Alibaba / Qwen
Model provider for Qwen-family public benchmark rows.
13models123results
08Z.ai
Model provider for GLM-family public benchmark rows.
7models107results
09Moonshot AI
Model provider for Kimi-family public benchmark rows.
6models108results
10Baidu
Model provider for ERNIE public benchmark rows.
1models2results
11Mistral AI
Model provider for Mistral-family public benchmark rows.
2models26results
12Cohere
Model provider for Command-family public benchmark rows.
1models1results
13MiniMax
Model provider for MiniMax public benchmark rows.
4models50results
14DeepReinforce
Model provider for DeepReinforce (Ornith-family) public benchmark rows.
1models4results
15Manus
Provider for the Manus agent product's public benchmark rows.
1models1results
16Agent systems
Source-reported agent or scaffold entries where the benchmark row is not a single base model.
31models39results
17Microsoft AI
Model provider for Microsoft AI (MAI) public benchmark rows.
2models9results
18NVIDIA
Model provider for Nemotron-family public benchmark rows.
3models28results
19ByteDance
Model provider for Doubao / Seed public benchmark rows.
2models7results
20Tencent
Model provider for Hunyuan public benchmark rows.
1models1results
21Xiaomi
Model provider for MiMo-family public benchmark rows.
4models21results
22Amazon
Model provider for Amazon Nova public benchmark rows.
2models13results
23Allen Institute for AI
Model provider for OLMo fully-open public benchmark rows.
1models2results
24LG AI Research
Model provider for EXAONE-family public benchmark rows.
1models4results
25StepFun
Model provider for Step-family public benchmark rows.
2models6results
26Upstage
Model provider for Solar-family public benchmark rows.
1models3results
27Reka AI
Model provider for Reka-family public benchmark rows.
1models3results
28IBM
Model provider for Granite-family public benchmark rows.
1models1results
29AI21 Labs
Model provider for Jamba-family public benchmark rows.
1models5results
30Nex AGI
Model provider for the Nex-N2 agentic-model public benchmark rows (open-weight, post-trained on Qwen3.5).
2models14results
31Cognition
Maker of Devin and the SWE-family coding models (Windsurf/Devin); creator of the FrontierCode benchmark.
3models6results
32Sakana AI
Provider of the Fugu multi-agent system, delivered as a single API that dynamically orchestrates multiple underlying LLMs.
2models14results
33Thinking Machines Lab
Model provider for Inkling-family open-weights public benchmark rows.
2models19results
34Cursor
Maker of the Cursor editor and the Composer coding models; creator of the CursorBench benchmark.
1models1results
35MBZUAI
Model provider for K2-family open-weights public benchmark rows.
1models1results
36Dots Studio
Model provider for dots-family open-weights public benchmark rows.
1models7results