Open Models Almanac

The models · 46 open-weight · 24 labs

46 open-weight models, newest first.

The open-weight models the almanac covers. Entry is by the OpenRouter weekly token ranking plus explicit additions; a model leaving the ranking keeps its page, because the methods its documentation described do not expire.

Sort by any column. “Methods” counts what a model’s sources describe, so it measures how openly a lab writes about its model — a 200-method report and a 17-method model card can describe equally capable models. Sizes, context, attention family and licence are recorded by hand from the model card and config.

Mistral-Large-4Mistral AI2026-10-29—~1T~52B1Msparse (indexer)1887
LFM3-1.7BLiquid AI2026-10-27—1.73Bdense256Kconv hybrid1645
MiniMax-M3.1-FlashMiniMax2026-10-19#6~428B~23B1Msparse (indexer)113
DeepSeek-V4.1-Flash-1017DeepSeek2026-10-17#1552B16B1Msparse (indexer)123
NVIDIA-Nemotron-3.5-Super-VL-120B-A12BNVIDIA2026-10-15—120B12B1MMamba hybrid2777
Step-5-PreviewStepFun2026-10-08#5600B27B1Msliding/global hybrid274
Kolibri-1Aleph Alpha2026-10-03—78B3.46B256Ksliding/global hybrid4104
FLUX 3 ActionBlack Forest Labs2026-09-23—7B7B—243
MiMo-V2.6-FlashXiaomi2026-09-22#10309B15B1Msliding/global hybrid1365
MiMo-V2.6-ProXiaomi2026-09-22—1.02T42B1Msliding/global hybrid1425
DeepSeek-V4.1-FlashDeepSeek2026-09-10#3552B16B1Msparse (indexer)1054
Hy4-previewTencent (Hunyuan)2026-08-28#4770B49B1Msparse (indexer)357
GLM-5.3-FlashZ.ai (Zhipu AI)2026-08-26#2320B18B1Mlinear hybrid335
Qwen3.8-Flash-NextAlibaba (Qwen)2026-08-26—125B6B256Klinear hybrid1217
DeepSeek-V4-Flash-Vision-ExpDeepSeek2026-08-21—284B13B1Msparse (indexer)253
GLM-5.3Z.ai (Zhipu AI)2026-08-16#7~753B~40B1Msparse (indexer)466
DeepSeek-V4-Pro-0813DeepSeek2026-08-13—1.6T49B1Msparse (indexer)133
NVIDIA-Nemotron-3.5-Lightning-30B-A3BNVIDIA2026-08-11—30B3B1MMamba hybrid2848
DeepSeek-V4-Flash-0731DeepSeek2026-07-31#8284B13B1Msparse (indexer)174
Xiaomi-Robotics-1Xiaomi2026-07-28————463
Laguna-S-2.1Poolside2026-07-21—118B8B1Msliding/global hybrid1638
Kimi K3Moonshot AI2026-07-16—2.8T104B1Mlinear hybrid18510
InklingThinking Machines Lab2026-07-15—975B41B1Msliding/global hybrid7211
LingBot-VLA 2.0Robbyant (Ant Group)2026-07-07—6.4B6.4B—915
Hy3Tencent (Hunyuan)2026-07-06#9295B21B256Kfull (GQA/MHA)377
GLM-5.2Z.ai (Zhipu AI)2026-06-16—~753B~40B1Msparse (indexer)685
NVIDIA-Nemotron-3-Ultra-550B-A55BNVIDIA2026-06-04#11550B55B1MMamba hybrid2057
MiniMax-M3MiniMax2026-06-01—~428B~23B1Msparse (indexer)817
Step-3.7-FlashStepFun2026-05-29—~196B~11B256Ksliding/global hybrid436
MolmoAct 2Allen Institute for AI2026-05-04—5.4B5.4B—1653
DeepSeek-V4-FlashDeepSeek2026-04-24#12284B13B1Msparse (indexer)824
DeepSeek-V4-ProDeepSeek2026-04-24—1.6T49B1Msparse (indexer)975
MiMo-V2.5Xiaomi2026-04-22—310B15B1Msliding/global hybrid768
MiMo-V2.5-ProXiaomi2026-04-22—1.02T42B1Msliding/global hybrid426
GR00T N1.7NVIDIA2026-04-17—3.1B3.1B—824
VLA Foundry (Foundry-VLA-1.7B)Toyota Research Institute2026-04-17—1.7B1.7B—683
Qwen3.6-35B-A3BAlibaba (Qwen)2026-04-15—35B3B256Klinear hybrid466
Gemma 4 31BGoogle DeepMind2026-04-02—30.7Bdense256Ksliding/global hybrid7910
Alpamayo 1.5 / 2 SuperNVIDIA2026-03-03—10B10B—612
Qwen3.5-397B-A17BAlibaba (Qwen)2026-02-16—397B17B256Klinear hybrid1008
DeepSeek-V3.2DeepSeek2025-12-01—671B37B160Ksparse (indexer)684
X-VLAX-VLA authors2025-11-04—0.9B0.9B—502
π0.5Physical Intelligence2025-09-09—3.6B3.6B—1075
gpt-oss-120bOpenAI2025-08-05—117B5.1B128Ksliding/global hybrid846
SmolVLAHugging Face2025-06-03—0.45B0.45B—623
OpenVLA / OpenVLA-OFTOpenVLA project (Stanford et al.)2024-06-10—7.5B7.5B—733

Rank is the model's position among open-weight models in OpenRouter's token ranking for the week of 2026-10-26, the same week for every model; — means outside that week's top 12. Params and Active are total and per-token parameters; hover an attention family for its detail.