Llama
15 models14 ranked conversions408.5K 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 | DeepSeek-R1-Distill-Llama-8B deepseek-ai | simmessa | 8B | 0.0015 | 3/1003% | 98.4 | BestParetoGGUFSAFETENSORS | 24.6K |
| 2 | Llama-Opus-Z8 Daemontatox | ChiKoi7 | 8B | 0.0071 | 3/1003% | 98.2 | BestParetoGGUFSAFETENSORS | 8K |
| 3 | TinyLlama-1.1B-Chat-v1.0 TinyLlama | rainmana | 1.1B | 0.024 | 2/1002% | 97.8 | BestParetoSAFETENSORS | 123 |
| 4 | Llama-SEA-LION-v3-8B-IT aisingapore | MagicalAlchemist | 8B | 0.031 | 9/1009% | 94.0 | BestParetoGGUFSAFETENSORS | 16.3K |
| 5 | Stylizer-V2-LLaMa-70B TareksGraveyard | Nabbers1999 | 70B | 0.042 | 8/1008% | 94.0 | BestParetoSAFETENSORS | 279 |
| 6 | DeepSeek-R1-Distill-Llama-70B deepseek-ai | Total04 | 70B | 0.033 | 16/10016% | 90.4 | BestParetoGGUFSAFETENSORS | 163.1K |
| 7 | Llama-2-7b-chat-hf meta-llama | HarethahMo | 7B | 0.0002 | 39/10039% | 80.5 | BestParetoGGUFNVFP4SAFETENSORS | 3.9K |
All models
DeepSeek-R1-Distill-Llama-8B8BLlama-Opus-Z88BTinyLlama-1.1B-Chat-v1.01.1BLlama-SEA-LION-v3-8B-IT8BStylizer-V2-LLaMa-70B70BDeepSeek-R1-Distill-Llama-70B70BLlama-2-7b-chat-hf7BHeretic-Goppa-LogiLlama—Heretic-Llama-Deepsync-1B1BHeretic-Llama-Express.1—Heretic-Reasoning-Llama-1b-v0.11BLlama-Poro-2-8B-Long-Instruct8BLlama-xLAM-2-8b-fc-r8Bllamas-heretic—Total04-DeepSeek-R1-Distill-Llama-70B-heretic70B
Unmeasured models
Heretic-Goppa-LogiLlama, Heretic-Llama-Deepsync-1B, Heretic-Llama-Express.1, Heretic-Reasoning-Llama-1b-v0.1, Llama-Poro-2-8B-Long-Instruct, Llama-xLAM-2-8b-fc-r, llamas-heretic, Total04-DeepSeek-R1-Distill-Llama-70B-heretic — conversions exist but their publishers did not report KL and refusal numbers, so they are not ranked.