mannix / deepseek-coder-v2-lite-instruct

An open-source Mixture-of-Experts code language model that achieves performance comparable to GPT4-Turbo in code-specific tasks.

555 Pulls Updated 2 months ago

23 Tags

latest

a3f7e2de4903 • 8.9GB • 3 months ago

fp16

efabe10c99de • 31GB • 2 months ago

iq2_s

e64344c032f0 • 6.0GB • 3 months ago

iq2_xs

b74d58868e20 • 6.0GB • 3 months ago

iq3_s

307e14eae256 • 7.5GB • 3 months ago

iq3_xs

8f4eb22b4880 • 7.1GB • 3 months ago

iq3_xxs

aa07d2877393 • 7.0GB • 3 months ago

iq4_nl

6a3530175855 • 8.9GB • 3 months ago

iq4_xs

06077e3cbc91 • 8.6GB • 3 months ago

q2_k

71002b284f0c • 6.4GB • 3 months ago

q3_k_l

92855e4e3b49 • 8.5GB • 3 months ago

q3_k_m

994a6e62c235 • 8.1GB • 3 months ago

q3_k_s

f917ef1b196e • 7.5GB • 3 months ago

q4_0

a3f7e2de4903 • 8.9GB • 3 months ago

q4_1

b84e1608422b • 9.9GB • 3 months ago

q4_k_m

6171206208d0 • 10GB • 3 months ago

q4_k_s

40644ffb404d • 9.5GB • 3 months ago

q5_0

1b8598c3d51f • 11GB • 3 months ago

q5_1

729988f9f073 • 12GB • 3 months ago

q5_k_m

aa6f4f5b69f0 • 12GB • 3 months ago

q5_k_s

b80760d1c9dc • 11GB • 3 months ago

q6_k

78582e0717cc • 14GB • 3 months ago

q8_0

f2a5c425a368 • 17GB • 3 months ago