Skip to content
AI Atlas
Families

Model families

Families group canonical releases (Llama 3.1, Qwen3, Claude…). Counts, releases and parameter ranges are aggregated from the members' sourced claims; benchmark ranks are the best rank of any member in the benchmark's primary group.

197 families

Reset
Model families
FamilyOrganizationModelsFirst · latest releaseParametersModalitiesLicencesBest benchmark ranks
Claude 35Anthropic2 · GPQA Diamond #278AA Intelligence Index #292Humanity's Last Exam #415
Claude 4Anthropic2 · SWE-bench Verified #13Terminal-Bench #79AA Intelligence Index #119
Claude 4.5Anthropic2 · SWE-bench Multilingual #8SWE-bench Verified #9
CogitoDeep Cogito2 · Terminal-Bench #137IFBench #162GPQA Diamond #171
DALL·EOpenAI2 · no results
EXAONE 4.0LG AI Research2 · Humanity's Last Exam #177GPQA Diamond #194Terminal-Bench #222
Embedding 3OpenAI2 · no results
GLMZ.ai (Zhipu AI)29 Dec 2025 · 30 Jan 20261.33B – 2.26BMIT 2no results
GLM4Z.ai (Zhipu AI)24 Jun 2024 · 4 Jun 20249.4BOther 1no results
GLM5.2Z.ai (Zhipu AI)216 Jun 2026 · 16 Jun 2026753.3BtextMIT 1τ²-bench #1Terminal-Bench #12LiveBench Coding #16
GLM5.3Z.ai (Zhipu AI)225 Aug 2026 · 25 Aug 2026321.3B – 753.3BimagetextvideoOther 1MIT 1SciCode #6AA Intelligence Index #7LiveBench Agentic Coding #13
GPT 4.1106OpenAI2 · no results
GPT 5.3OpenAI224 Feb 2026 · 24 Feb 2026documentimagetextTerminal-Bench #9Humanity's Last Exam #21GPQA Diamond #27
GPT 6OpenAI24 Sept 2026 · 4 Sept 2026documentimagetextGPQA Diamond #1LiveBench Data Analysis #1LiveBench Reasoning #1
Gemini 1.0Google2 · Humanity's Last Exam #373AA Intelligence Index #395GPQA Diamond #437
Granite 3.0IBM22 Oct 2024 · 2 Oct 20248.17BApache-2.0 1no results
Grok 1xAI217 Mar 2024 · 17 Mar 2024Apache-2.0 1AA Intelligence Index #373
Grok 4SpaceXAI2 · Aider polyglot #5Aider polyglot — well-formed responses #27Terminal-Bench #44
Hermes 3Nous Research216 Aug 2024 · 18 Aug 2024textGPQA Diamond #374AA Intelligence Index #383Humanity's Last Exam #393
Hermes 4Nous Research218 Aug 2025 · 26 Aug 202570.5BtextLlama-3-Community 1Terminal-Bench #159Humanity's Last Exam #187GPQA Diamond #206
InternLM2.1InternLM (Shanghai AI Laboratory)230 Jan 2024 · 27 Jun 20241.7BOther 2no results
Jamba 1.5AI21 Labs2 · Humanity's Last Exam #303GPQA Diamond #359AA Intelligence Index #382
Jamba 1.6AI21 Labs2 · Humanity's Last Exam #360GPQA Diamond #379AA Intelligence Index #386
Jamba 1.7AI21 Labs2 · IFBench #240Terminal-Bench #240τ²-bench #277
Llama 3.1Meta AI223 Jul 2024 · 23 Jul 20248.03B – 70.5BtextLlama-3.1-Community 2no results
Llama 3.3NVIDIA2 · IFBench #160τ²-bench #209Terminal-Bench #233
Llama3Sao10K213 Aug 2024 · 13 Aug 2024textno results
OLMo 2Allen Institute for AI2 · IFBench #216Terminal-Bench #279Humanity's Last Exam #291
OLMo 2.0325Allen Institute for AI212 Mar 2025 · 12 Mar 202532.2BApache-2.0 1no results
PixtralMistral AI2 · MMMU-Pro #128τ²-bench #170IFBench #244
QwenQwen21 Feb 2025 · 8 Sept 2025textno results
Seed 1.6ByteDance223 Dec 2025 · 23 Dec 2025imagetextvideono results
Sora 2OpenAI2 · no results
Veo 3.1Google2 · no results
Aya 101Cohere18 Feb 2024 · 8 Feb 202412.9BApache-2.0 1no results
Aya 23Cohere119 May 2024 · 19 May 20248.03BCC-BY-NC-4.0 1no results
Claude 2Anthropic1 · SWE-bench (full test split) #2SWE-bench Lite #2GPQA Diamond #394
Claude 21Anthropic1 · Humanity's Last Exam #398AA Intelligence Index #409GPQA Diamond #413
Codestral 24.05Mistral AI1 · no results
Codestral 25.01Mistral AI1 · Aider polyglot — well-formed responses #1Aider polyglot #57
Codestral 2508Mistral AI11 Aug 2025 · 1 Aug 2025documenttextno results
Devstral 2Mistral AI19 Dec 2025 · 9 Dec 2025documenttextSWE-bench Verified #28SciCode #105Terminal-Bench #124
ERNIE 5.0Baidu1 · τ²-bench #86MMMU-Pro #92Terminal-Bench #100
EXAONELG AI Research1 · IFBench #89τ²-bench #112Terminal-Bench #113
EXAONE 2.0LG AI Research1 · SciCode #74GPQA Diamond #120AA Intelligence Index #126
EXAONE 4.5LG AI Research1 · MMMU-Pro #83τ²-bench #105IFBench #111
Embedding 2Salesforce114 Jun 2024 · 14 Jun 20247.11BCC-BY-NC-4.0 1no results
FalconTII UAE1 · IFBench #127Humanity's Last Exam #184τ²-bench #202
GLM4.1Z.ai (Zhipu AI)128 Jun 2025 · 28 Jun 202510.3BMIT 1no results
GLM5.1Z.ai (Zhipu AI)13 Apr 2026 · 3 Apr 2026753.9BtextMIT 1τ²-bench #9IFBench #21Terminal-Bench #27

Family canonical: true = model_family entity; false = legacy attributes.family label awaiting canonicalisation. benchmark_best = best rank of any member in each benchmark's primary group.