StarCoder2 is the next generation of transparently trained open code LLMs that comes in three sizes: 3B, 7B and 15B parameters.
3B
7B
16B
60.4K Pulls Updated 2 weeks ago
instruct
9.1GB
15b-instruct
9.1GB
15b-q4_0
9.1GB
15b-q4_1
10GB
15b-q5_0
11GB
15b-q5_1
12GB
15b-q8_0
17GB
15b-q2_K
6.2GB
15b-q3_K_S
7.0GB
15b-q3_K_M
8.1GB
15b-q3_K_L
9.0GB
15b-q4_K_S
9.3GB
15b-q4_K_M
9.9GB
15b-q5_K_S
11GB
15b-q5_K_M
11GB
15b-q6_K
13GB
15b-fp16
32GB
15b-instruct-q4_0
9.1GB
15b-instruct-v0.1-q4_0
9.1GB
15b-instruct-v0.1-q4_1
10GB
15b-instruct-v0.1-q5_0
11GB
15b-instruct-v0.1-q5_1
12GB
15b-instruct-v0.1-q8_0
17GB
15b-instruct-v0.1-q2_K
6.2GB
15b-instruct-v0.1-q3_K_S
7.0GB
15b-instruct-v0.1-q3_K_M
8.0GB
15b-instruct-v0.1-q3_K_L
9.0GB
15b-instruct-v0.1-q4_K_S
9.2GB
15b-instruct-v0.1-q4_K_M
9.9GB
15b-instruct-v0.1-q5_K_S
11GB
15b-instruct-v0.1-q5_K_M
11GB
15b-instruct-v0.1-q6_K
13GB
15b-instruct-v0.1-fp16
32GB
7b-q4_0
4.0GB
7b-q4_1
4.5GB
7b-q5_0
4.9GB
7b-q5_1
5.4GB
7b-q8_0
7.6GB
7b-q2_K
2.7GB
7b-q3_K_S
3.1GB
7b-q3_K_M
3.6GB
7b-q3_K_L
4.0GB
7b-q4_K_S
4.1GB
7b-q4_K_M
4.4GB
7b-q5_K_S
4.9GB
7b-q5_K_M
5.1GB
7b-q6_K
5.9GB
7b-fp16
14GB
3b-q4_0
1.7GB
3b-q4_1
1.9GB
3b-q5_0
2.1GB
3b-q5_1
2.3GB
3b-q8_0
3.2GB
3b-q2_K
1.1GB
3b-q3_K_S
1.3GB
3b-q3_K_M
1.5GB
3b-q3_K_L
1.7GB
3b-q4_K_S
1.7GB
3b-q4_K_M
1.8GB
3b-q5_K_S
2.1GB
3b-q5_K_M
2.2GB
3b-q6_K
2.5GB
3b-fp16
6.1GB