|
Type,Name,MT-Bench (coming soon!),Average,ARC,HellaSwag,MMLU,TruthfulQA,Winogrande,GSM8K,Architecture,Model Size,Base Model |
|
πΆ SFT,recursal/EagleX_1-7T_Chat,,47.36,47.87,73.38,34.73,42.41,73.09,12.66,RWKV-5,7B,recursal/EagleX_1-7T |
|
π’ Pretrained,recursal/EagleX_1-7T,,46.84,47.10,73.03,34.91,40.18,73.24,12.59,RWKV-5,7B,base |
|
πΆ SFT,EleutherAI/Hermes-RWKV-v5-7B,,46.42,48.04,72.88,31.66,42.02,69.53,14.40,RWKV-5,7B,RWKV/v5-Eagle-7B |
|
π’ Pretrained,RWKV/v5-Eagle-7B,,45.15,47.61,71.66,31.01,40.62,70.80,09.17,RWKV-5,7B,base |
|
π¦ RLHF/DPO,jondurbin/bagel-dpo-2.8b-v0.2,,43.89,43.43,70.47,37.10,43.22,68.67,00.45,Mamba,3B,jondurbin/bagel-2.8b-v0.2 |
|
πΆ SFT,jondurbin/bagel-2.8b-v0.2,,43.47,41.55,69.87,35.83,43.79,67.88,01.90,Mamba,3B,state-spaces/mamba-2.8b-slimpj |
|
πΆ SFT,RWKV/rwkv-raven-14b,,42.09,44.62,71.25,25.92,41.93,66.69,02.12,RWKV-4,13B,RWKV/rwkv-4-14b-pile |
|
π¦ RLHF/DPO,EleutherAI/Hermes-mamba-2.8b-slimpj-cDPO,,41.77,42.15,71.84,27.50,37.69,67.09,04.32,Mamba,3B,EleutherAI/Hermes-mamba-2.8b-slimpj |
|
πΆ SFT,EleutherAI/Hermes-mamba-2.8b-slimpj,,41.65,41.64,71.46,27.65,37.31,66.85,05.00,Mamba,3B,state-spaces/mamba-2.8b-slimpj |
|
π¦ RLHF/DPO,xiuyul/mamba-2.8b-zephyr,,41.59,44.20,72.02,25.33,37.85,67.17,02.96,Mamba,3B,xiuyul/mamba-2.8b-ultrachat |
|
πΆ SFT,xiuyul/mamba-2.8b-ultrachat,,40.94,43.26,71.19,25.28,36.69,66.54,02.65,Mamba,3B,state-spaces/mamba-2.8b-slimpj |
|
π’ Pretrained,state-spaces/mamba-2.8b-slimpj,,40.68,43.43,71.38,26.19,34.35,66.38,02.35,Mamba,3B,base |
|
πΆ SFT,Trelis/mamba-2.8b-slimpj-chat-4k,,40.20,41.72,70.18,25.76,36.08,66.61,00.83,Mamba,3B,state-spaces/mamba-2.8b-slimpj |
|
π’ Pretrained,RWKV/rwkv-4-14b-pile,,39.92,44.45,71.07,26.12,32.04,65.43,00.38,RWKV-4,13B,base |
|
πΆ SFT,clibrain/mamba-2.8b-chat-no_robots,,39.48,41.55,68.02,26.00,35.81,63.30,02.20,Mamba,3B,state-spaces/mamba-2.8b |
|
πΆ SFT,clibrain/mamba-2.8b-instruct-openhermes,,39.20,40.96,65.61,24.62,36.60,63.46,03.94,Mamba,3B,state-spaces/mamba-2.8b |
|
π’ Pretrained,state-spaces/mamba-2.8b,,38.94,39.93,66.47,26.09,35.72,64.09,01.36,Mamba,3B,base |
|
πΆ SFT,EleutherAI/Hermes-mamba-2.8b,,38.93,37.46,66.25,25.13,36.48,64.17,04.09,Mamba,3B,state-spaces/mamba-2.8b |
|
πΆ SFT,havenhq/mamba-chat,,38.93,40.96,66.40,25.34,36.36,62.83,01.67,Mamba,3B,state-spaces/mamba-2.8b |
|
πΆ SFT,RWKV/rwkv-raven-7b,,38.55,39.42,66.48,23.64,38.56,62.90,00.30,RWKV-4,7B,RWKV/rwkv-4-7b-pile |
|
π’ Pretrained,RWKV/rwkv-4-7b-pile,,37.95,39.68,66.31,24.96,33.65,62.35,00.76,RWKV-4,7B,base |
|
π’ Pretrained,RWKV/rwkv-4-world-7b,,37.79,38.65,65.59,25.94,34.20,62.35,00.00,RWKV-4,7B,base |
|
π’ Pretrained,RWKV/rwkv-5-world-3b,,37.75,38.82,62.74,25.55,36.22,63.14,00.02,RWKV-5,3B,base |
|
π’ Pretrained,state-spaces/mamba-1.4b,,36.15,35.15,59.19,25.21,35.21,61.09,01.06,Mamba,1.5B,base |
|
π’ Pretrained,RWKV/rwkv-5-world-1b5,,36.10,36.60,55.20,25.97,38.74,58.96,01.14,RWKV-5,1.5B,base |
|
π’ Pretrained,RWKV/rwkv-4-world-3b,,36.04,37.12,58.95,25.06,35.92,59.19,00.00,RWKV-4,3B,base |
|
πΆ SFT,RWKV/rwkv-raven-3b,,35.81,36.69,59.78,24.87,35.60,57.46,00.45,RWKV-4,3B,RWKV/rwkv-4-3b-pile |
|
π’ Pretrained,RWKV/rwkv-4-3b-pile,,35.25,36.01,59.66,24.67,32.14,58.33,00.68,RWKV-4,3B,base |
|
πΆ SFT,RWKV/rwkv-raven-1b5,,33.56,31.83,52.60,25.96,37.09,53.91,00.00,RWKV-4,1.5B,RWKV/rwkv-4-1b5-pile |
|
π’ Pretrained,RWKV/rwkv-4-1b5-pile,,33.25,31.83,52.25,25.77,35.80,53.83,00.00,RWKV-4,1.5B,base |
|
π Running,EleutherAI/Hermes-RWKV-v5-3B-HF,,,,,,,,,RWKV-5,3B,RWKV/rwkv-5-world-3b |
|
π Running,EleutherAI/Hermes-RWKV-v4-3B,,,,,,,,,RWKV-4,3B,RWKV/rwkv-4-3b-pile |
|
π Running,RWKV/rwkv-4-world-1b5,,,,,,,,,RWKV-4,1.5B,base |
|
β³ Pending,TimeMobius/Mobius-mega-12B-128k-base,,,,,,,,,RWKV-5,13B,base |
|
β³ Pending,TimeMobius/Mobius-Chat-12B-128k,,,,,,,,,RWKV-5,13B,TimeMobius/Mobius-mega-12B-128k-base |
|
β³ Pending,TimeMobius/Mobius-RWKV-Chat-12B-128k-Sacred-Timeline,,,,,,,,,RWKV-5,13B,TimeMobius/Mobius-mega-12B-128k-base |
|
β³ Pending,togethercomputer/StripedHyena-Hessian-7B,,,,,,,,,StripedHyena,7B,base |
|
β³ Pending,togethercomputer/StripedHyena-Nous-7B,,,,,,,,,StripedHyena,7B,togethercomputer/StripedHyena-Hessian-7B |