Skip to content
AI Atlas
Open model frontier

Open models

Every canonical model whose weights can be downloaded, described by measurable properties — licence permissions, parameters, context, release, benchmark ranks, providers, estimated local fit. Openness is a set of observed dimensions here, never a score.

950 downloadable models

Open sourceWeights and code published under an OSI-approved licence that allows commercial use and derivatives.

1

Open weightsWeights downloadable under a permissive or Creative Commons licence allowing commercial use; code or data may be missing.

869

Restricted weightsWeights downloadable, but the licence restricts commercial use, hosting, derivatives or field of use (community, research and RAIL licences).

80

Downloadable totalUniverse = canonical models whose weights can be downloaded (open-weights, open-source, restricted-weights). hardware_fit values are ESTIMATES (see /methodology); best_results are ranks inside each benchmark's primary comparability group — no composite score.

950

New in 30 daysDownloadable models with a release date inside the last 30 days (release_date claim; models without one are not counted).

269

Openness explorer

Categories and licence permissions

Left: the downloadable universe by openness category. Right: what the licences on this page allow, counted per category — allowed · restricted · unknown.

Top licences: Apache License 2.0 (275) · Other (unclassified licence) (115) · MIT License (92) · Creative Commons Attribution-NonCommercial 4.0 (22) · Apple Sample Code / ML Research License (15) · Llama 3.2 Community License (12)

Category · this pageCommercial useRedistributionDerivativesHosting
Open source n=0————
Open weights n=469037903790379037
restricted-weights n=4400400400040

Counts read: allowed · restricted · unknown / unclassified — over the 50 models on this page, from each licence's stated terms (ontology).

Downloadable models

950 models · sorted by best benchmark rank

Permissions, in order:commercial use· redistribution· derivatives· hosting allowed restricted unknown

ModelLicence · permissionsParamsContextReleaseBest resultsProvidersFit 64 GB @4bit · 128 GB @8bit Estimated
lfm2-5-1-2b-thinkingOpen weightsLiquid AI—unclassified—32K—ifbench#187tau2-bench#253humanitys-last-exam#2620——
llama-nemotron-super-49b-v1-5Open weightsNVIDIA—unclassified—128K—gpqa-diamond#189tau2-bench#199terminal-bench#2050——
Olmo-3-7B-ThinkOpen weightsAllen Institute for AIApache-2.07.3B65.5K18 Nov 2025ifbench#190terminal-bench#262humanitys-last-exam#2700✓ fits4.7 GB4bit✓ fits8.9 GB8bit
mistral-large-2Open weightsMistral AI—unclassified—128K—tau2-bench#191terminal-bench#196ifbench#2710——
devstral-smallOpen weightsMistral AI—unclassified—256K—terminal-bench#196tau2-bench#198ifbench#2410——
granite-4.1-8bOpen weightsIBMApache-2.08.79B131.1K6 Apr 2026tau2-bench#202ifbench#212terminal-bench#2760✓ fits5.6 GB4bit✓ fits10.6 GB8bit
llama-3-1-nemotron-ultra-253b-v1-reasoningOpen weightsNVIDIA—unclassified—128K—gpqa-diamond#205ifbench#214humanitys-last-exam#2350——
llama-3-3-nemotron-super-49bOpen weightsNVIDIA—unclassified—128K—ifbench#205tau2-bench#209gpqa-diamond#2580——
Hermes-4-70Brestricted-weightsNous ResearchLlama-3-Community70.5B128K18 Aug 2025terminal-bench#208humanitys-last-exam#219gpqa-diamond#2280✓ fits41.1 GB4bit✓ fits81.6 GB8bit
llama-3-1-nemotron-instruct-70bOpen weightsNVIDIA—unclassified—128K—terminal-bench#208tau2-bench#232ifbench#2740——
magistral-smallOpen weightsMistral AI—unclassified—40K—terminal-bench#208tau2-bench#210humanitys-last-exam#2300——
olmo-2-32bOpen weightsAllen Institute for AI—unclassified—4.1K—ifbench#215terminal-bench#276tau2-bench#3020——
qwen3-1.7b-instructOpen weightsAlibaba Group—unclassified—32K—tau2-bench#217terminal-bench#276ifbench#2910——
Phi 4Open weightsMicrosoft · textMIT14.7B16.4K11 Dec 2024terminal-bench#220gpqa-diamond#295tau2-bench#3021from $0.14 out✓ fits8.9 GB4bit✓ fits17.4 GB8bit
llama-3-instruct-70bOpen weightsMeta AI—unclassified—8.19K—ifbench#222terminal-bench#262tau2-bench#3020——
g9v3-3bOpen weightsAI9Stars—unclassified—131.1K—artificial-analysis-intelligence-index#223humanitys-last-exam#344gpqa-diamond#3520——
qwen3-8b-instructOpen weightsAlibaba Group—unclassified—32.8K—tau2-bench#223terminal-bench#237ifbench#2830——
Command AOpen weightsCohere · text—unclassified—256K13 Mar 2025ifbench#229terminal-bench#262tau2-bench#2701from $10 out——
NVIDIA-Nemotron-Nano-9B-v2Open weightsNVIDIA—unclassified—131.1K—tau2-bench#229terminal-bench#253ifbench#2880——
Llama-3.1-70BOpen weightsMeta AI—unclassified—128K—terminal-bench#231ifbench#243tau2-bench#2700——
granite-4-0-nano-1bOpen weightsIBM—unclassified—128K—tau2-bench#234terminal-bench#276humanitys-last-exam#3220——
granite-4-0-h-smallOpen weightsIBM—unclassified—128K—terminal-bench#237tau2-bench#262ifbench#2660——
granite-4.1-3bOpen weightsIBMApache-2.03.4B131.1K6 Apr 2026terminal-bench#237ifbench#250tau2-bench#2530✓ fits2.5 GB4bit✓ fits4.4 GB8bit
jamba-1-7-largeOpen weightsAI21 Labs—unclassified—256K—terminal-bench#237ifbench#238tau2-bench#2770——
sarvam-m-reasoningOpen weightsSarvam—unclassified—32.8K—terminal-bench#237ifbench#262tau2-bench#3020——
R1 Distill Llama 70BOpen weightsDeepSeek · text—unclassified—8.19K23 Jan 2025tau2-bench#239terminal-bench#253artificial-analysis-intelligence-index#2891from $0.80 out——
hermes-4-llama-3-1-70bOpen weightsNous Research—unclassified—128K—tau2-bench#240terminal-bench#276ifbench#2820——
Llama-3.2-3Brestricted-weightsMeta AILlama-3.2-Community3.21B128K18 Sept 2024tau2-bench#245humanitys-last-exam#290ifbench#2970✓ fits2.4 GB4bit✓ fits4.2 GB8bit
Qwen3-0.6BOpen weightsQwenApache-2.0751.6M32K27 Apr 2025tau2-bench#245terminal-bench#276humanitys-last-exam#2770✓ fits0.9 GB4bit✓ fits1.4 GB8bit
exaone-4-0-1-2bOpen weightsLG AI Research—unclassified—64K—tau2-bench#249humanitys-last-exam#270terminal-bench#2760——
solar-miniOpen weightsUpstage—unclassified—4.1K—tau2-bench#250artificial-analysis-intelligence-index#3690——
granite-4-0-h-350mOpen weightsIBM—unclassified—32.8K—humanitys-last-exam#252tau2-bench#272terminal-bench#2760——
DeepSeek-R1-0528-Qwen3-8BOpen weightsDeepSeekMIT8.19B32.8K29 May 2025terminal-bench#253humanitys-last-exam#272gpqa-diamond#2730✓ fits5.2 GB4bit✓ fits9.9 GB8bit
Granite 4.0 MicroOpen weightsIBM · text—unclassified—131K20 Oct 2025terminal-bench#253tau2-bench#282ifbench#3031from $0.112 out——
Mistral Small 3Open weightsMistral AI · text—unclassified—32.8K30 Jan 2025tau2-bench#253ifbench#295gpqa-diamond#3481from $0.08 out——
granite-4-0-h-nano-1bOpen weightsIBM—unclassified—128K—tau2-bench#253terminal-bench#276ifbench#2970——
Olmo-3-7B-InstructOpen weightsAllen Institute for AIApache-2.07.3B65.5K19 Nov 2025ifbench#255humanitys-last-exam#275terminal-bench#2760✓ fits4.7 GB4bit✓ fits8.9 GB8bit
Qwen3-4BOpen weightsQwenApache-2.04.02B32K27 Apr 2025tau2-bench#257ifbench#258gpqa-diamond#3150✓ fits2.8 GB4bit✓ fits5.1 GB8bit
Llama 3.1 8Brestricted-weightsMeta AILlama-3.1-Community8.03B128K14 Jul 2024terminal-bench#262tau2-bench#265ifbench#2830✓ fits5.1 GB4bit✓ fits9.7 GB8bit
gemma-3n-e2bOpen weightsGoogle—unclassified—32K—terminal-bench#262tau2-bench#302ifbench#3170——
lfm2-2-6bOpen weightsLiquid AI—unclassified—32.8K—terminal-bench#262tau2-bench#277humanitys-last-exam#2830——
ling-mini-2-0Open weightsinclusionAI—unclassified—131K—terminal-bench#262tau2-bench#279gpqa-diamond#3000——
qwen3-32b-instructOpen weightsAlibaba Group—unclassified—32.8K—ifbench#266gpqa-diamond#309artificial-analysis-intelligence-index#3190——
jamba-1-7-miniOpen weightsAI21 Labs—unclassified—258K—ifbench#269terminal-bench#276tau2-bench#2820——
DeepSeek-R1-Distill-Qwen-32BOpen weightsDeepSeekMIT32.8B128K20 Jan 2025gpqa-diamond#271artificial-analysis-intelligence-index#279ifbench#3120✓ fits19.3 GB4bit✓ fits38.2 GB8bit
qwen3-0.6b-instructOpen weightsAlibaba Group—unclassified—32K—tau2-bench#272terminal-bench#276humanitys-last-exam#3140——
Reka Flash 3Open weightsrekaai · text—unclassified—65.5K12 Mar 2025ifbench#275terminal-bench#276tau2-bench#3021from $0.20 out——
LFM2-1.2BOpen weightsLiquid AIOther1.17B32.8K10 Jul 2025terminal-bench#276humanitys-last-exam#279tau2-bench#2820✓ fits1.2 GB4bit✓ fits1.9 GB8bit
Llama-3.2-1Brestricted-weightsMeta AILlama-3.2-Community1.24B128K18 Sept 2024terminal-bench#276humanitys-last-exam#286tau2-bench#3020✓ fits1.2 GB4bit✓ fits1.9 GB8bit
apertus-70b-instructOpen weightsSwiss AI Initiative—unclassified—65.5K—terminal-bench#276tau2-bench#281humanitys-last-exam#2830——

Best results = rank inside each benchmark's primary comparability group; several dimensions are shown side by side and never combined. Fit columns are estimates (64 GB device at 4-bit, 128 GB at 8-bit, 8K context) — method on /methodology. unclassified = the raw licence label is not yet mapped in the ontology.

MethodUniverse = canonical models whose weights can be downloaded (open-weights, open-source, restricted-weights). hardware_fit values are ESTIMATES (see /methodology); best_results are ranks inside each benchmark's primary comparability group — no composite score. /methodology