Llama 3.1
21 models17 ranked conversions377.6K downloads
Best conversion per model
One row per base model. The full list of attempts is on each model page.
| # | Model | Author | Params | KL | Refusals | Score | Evidence | Downloads |
|---|---|---|---|---|---|---|---|---|
| 1 | Dolphin3.0-Llama3.1-8B dphn | anasali4151 | 8B | 0.0082 | 0/1000% | 99.6 | BestParetoGGUFSAFETENSORS | 24.7K |
| 2 | Llama-3.1-8B-Lexi-Uncensored-V2 Orenguteng | Spaceballs | 8B | 0.015 | 2/1002% | 98.3 | BestParetoSAFETENSORS | 214 |
| 3 | Llama-3.1-8B-Instruct meta-llama | askalgore | 8B | 0.017 | 3/1003% | 97.7 | BestParetoGGUFSAFETENSORS | 6.3K |
| 4 | Llama-3.1-8B-Instruct unsloth | ygee902 | 8B | 0.020 | 3/1003% | 97.5 | BestParetoSAFETENSORS | 5.9K |
| 5 | liberalis-cogitator-llama-3.1-8b-dpo Locutusque | MuXodious | 8B | 0.038 | 4/1004% | 96.2 | BestParetoGGUFSAFETENSORS | 7.7K |
| 6 | Llama-3.1-SuperNova-8B-Lite_TIES_with_Base Joseph717171 | ChiKoi7 | 8B | 0.053 | 3/1003% | 96.0 | BestParetoGGUFSAFETENSORS | 13K |
| 7 | LongWriter-llama3.1-8b zai-org | ChiKoi7 | 8B | 0.022 | 6/1006% | 95.9 | BestParetoGGUFSAFETENSORS | 6.3K |
| 8 | Llama-3.1-Nemotron-Nano-4B-v1.1 nvidia | hardcoreerik | 4B | 0.016 | 7/1007% | 95.7 | BestParetoGGUFSAFETENSORS | 6.7K |
| 9 | llama3.1-gutenberg-8B nbeerbower | ChiKoi7 | 8B | 0.061 | 3/1003% | 95.6 | BestParetoGGUFSAFETENSORS | 15K |
| 10 | Llama3.1-SuperHawk-8B Yuma42 | ChiKoi7 | 8B | 0.049 | 5/1005% | 95.2 | BestParetoGGUFSAFETENSORS | 9.4K |
| 11 | Llama3.1-IgneousIguana-8B Yuma42 | ChiKoi7 | 8B | 0.082 | 3/1003% | 94.7 | BestParetoGGUFSAFETENSORS | 9.5K |
| 12 | Llama-3.1-Hawkish-8B mukaj | ChiKoi7 | 8B | 0.059 | 8/1008% | 93.2 | BestParetoGGUFSAFETENSORS | 12.5K |
All models
Dolphin3.0-Llama3.1-8B8BLlama-3.1-8B-Lexi-Uncensored-V28BLlama-3.1-8B-Instruct8BLlama-3.1-8B-Instruct8Bliberalis-cogitator-llama-3.1-8b-dpo8BLlama-3.1-SuperNova-8B-Lite_TIES_with_Base8BLongWriter-llama3.1-8b8BLlama-3.1-Nemotron-Nano-4B-v1.14Bllama3.1-gutenberg-8B8BLlama3.1-SuperHawk-8B8BLlama3.1-IgneousIguana-8B8BLlama-3.1-Hawkish-8B8BHermes-3-Llama-3.1-8B8BLlama-3.1-70B-heretic-lora70.6BLlama-3.1-8B-Instruct-heretic8BLlama-3.1-8B-Instruct-heretic-V1-18BLlama-3.1-8B-Stheno-v3.4-Heretic8BLlama-3.1-Nemotron-8B-UltraLong-1M-Instruct_Abliterated8BMeta-Llama-3.1-8B-Instruct8BMeta-Llama-3.1-8B-Instruct-heretic8BToolACE-2.5-Llama-3.1-8B8B
Unmeasured models
Hermes-3-Llama-3.1-8B, Llama-3.1-70B-heretic-lora, Llama-3.1-8B-Instruct-heretic, Llama-3.1-8B-Instruct-heretic-V1-1, Llama-3.1-8B-Stheno-v3.4-Heretic, Llama-3.1-Nemotron-8B-UltraLong-1M-Instruct_Abliterated, Meta-Llama-3.1-8B-Instruct, Meta-Llama-3.1-8B-Instruct-heretic, ToolACE-2.5-Llama-3.1-8B — conversions exist but their publishers did not report KL and refusal numbers, so they are not ranked.