Downloads · 30 days
9
1% of all-time downloads
RichardErkhov/sydonayrex_-_Blackjack-Llama3-21B-gguf
sydonayrex_-_Blackjack-Llama3-21B-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
9
1% of all-time downloads
All-time downloads
850
Public
Repo size
243 GB
Likes
0
Public
Click a slice to open those files.
.gguf281 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
Blackjack-Llama3-21B - GGUF
| Name | Quant method | Size |
|---|---|---|
| Blackjack-Llama3-21B.Q2_K.gguf | Q2_K | 7.52GB |
| Blackjack-Llama3-21B.IQ3_XS.gguf | IQ3_XS | 8.34GB |
| Blackjack-Llama3-21B.IQ3_S.gguf | IQ3_S | 8.79GB |
| Blackjack-Llama3-21B.Q3_K_S.gguf | Q3_K_S | 8.74GB |
| Blackjack-Llama3-21B.IQ3_M.gguf | IQ3_M | 9.06GB |
| Blackjack-Llama3-21B.Q3_K.gguf | Q3_K | 9.69GB |
| Blackjack-Llama3-21B.Q3_K_M.gguf | Q3_K_M | 9.69GB |
| Blackjack-Llama3-21B.Q3_K_L.gguf | Q3_K_L | 10.52GB |
| Blackjack-Llama3-21B.IQ4_XS.gguf | IQ4_XS | 10.86GB |
| Blackjack-Llama3-21B.Q4_0.gguf | Q4_0 | 11.31GB |
| Blackjack-Llama3-21B.IQ4_NL.gguf | IQ4_NL | 11.43GB |
| Blackjack-Llama3-21B.Q4_K_S.gguf | Q4_K_S | 11.39GB |
| Blackjack-Llama3-21B.Q4_K.gguf | Q4_K | 12.01GB |
| Blackjack-Llama3-21B.Q4_K_M.gguf | Q4_K_M | 12.01GB |
| Blackjack-Llama3-21B.Q4_1.gguf | Q4_1 | 12.52GB |
| Blackjack-Llama3-21B.Q5_0.gguf | Q5_0 | 13.74GB |
| Blackjack-Llama3-21B.Q5_K_S.gguf | Q5_K_S | 13.74GB |
| Blackjack-Llama3-21B.Q5_K.gguf | Q5_K | 14.09GB |
| Blackjack-Llama3-21B.Q5_K_M.gguf | Q5_K_M | 14.09GB |
| Blackjack-Llama3-21B.Q5_1.gguf | Q5_1 | 14.95GB |
| Blackjack-Llama3-21B.Q6_K.gguf | Q6_K | 16.31GB |
| Blackjack-Llama3-21B.Q8_0.gguf | Q8_0 | 21.12GB |
base_model: sydonayrex/AI-Llama3-21B language:
The provided model is a multi-layered folded model, using multiple layers from the base Llama3 8B Instruct base, to increase its size to 21B parameters using mergekit. Rather than just using passthrough, task arithmetic was used. Further fine tuning was performed to ensure the model's weights and inference should be rebaselined.
q3_k_s GGUF :https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q3_K_S-GGUF
q4_k_m GGUF :https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q4_K_M-GGUF
q6_k GGUF: https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q6_K-GGUF
q8_0 GGUF: https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q8_0-GGUF
Only minor follow-up inference testing was performed after training.
This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.
Llama image generated by Meta AI.