deepseek-r1

DeepSeek's first-generation of reasoning models with comparable performance to OpenAI-o1, including six dense models distilled from DeepSeek-R1 based on Llama and Qwen.

1.5b 7b 8b 14b 32b 70b 671b

8.7M 2 weeks ago

28 Tags
latest
0a8c26691023 • 4.7GB • 2 weeks ago
1.5b
a42b25d8c10a • 1.1GB • 2 weeks ago
7b
0a8c26691023 • 4.7GB • 2 weeks ago
8b
28f8fd6cdc67 • 4.9GB • 2 weeks ago
14b
ea35dfe18182 • 9.0GB • 2 weeks ago
32b
38056bbcbb2d • 20GB • 2 weeks ago
70b
0c1615a8ca32 • 43GB • 2 weeks ago
671b
739e1b229ad7 • 404GB • 2 weeks ago
1.5b-qwen-distill-fp16
b3e50d21d5a9 • 3.6GB • 2 weeks ago
1.5b-qwen-distill-q4_K_M
a42b25d8c10a • 1.1GB • 2 weeks ago
1.5b-qwen-distill-q8_0
be5781997fd9 • 1.9GB • 2 weeks ago
14b-qwen-distill-fp16
1f1bddb1a1a2 • 30GB • 2 weeks ago
14b-qwen-distill-q4_K_M
ea35dfe18182 • 9.0GB • 2 weeks ago
14b-qwen-distill-q8_0
022efe288297 • 16GB • 2 weeks ago
32b-qwen-distill-fp16
141ef25faf00 • 66GB • 2 weeks ago
32b-qwen-distill-q4_K_M
38056bbcbb2d • 20GB • 2 weeks ago
32b-qwen-distill-q8_0
aed2818831dd • 35GB • 2 weeks ago
671b-fp16
3e2622881ef5 • 1.3TB • 2 weeks ago
671b-q8_0
ab9331000874 • 713GB • 2 weeks ago
70b-llama-distill-fp16
5c3319143a79 • 141GB • 2 weeks ago
70b-llama-distill-q4_K_M
0c1615a8ca32 • 43GB • 2 weeks ago
70b-llama-distill-q8_0
cc1ba146470d • 75GB • 2 weeks ago
7b-qwen-distill-fp16
fb45459f46e6 • 15GB • 2 weeks ago
7b-qwen-distill-q4_K_M
0a8c26691023 • 4.7GB • 2 weeks ago
7b-qwen-distill-q8_0
0bcb1414f90e • 8.1GB • 2 weeks ago
8b-llama-distill-fp16
b15636d30796 • 16GB • 2 weeks ago
8b-llama-distill-q4_K_M
28f8fd6cdc67 • 4.9GB • 2 weeks ago
8b-llama-distill-q8_0
0db4dcd4c434 • 8.5GB • 2 weeks ago