granite3-moe

The IBM Granite 1B and 3B models are the first mixture of experts (MoE) Granite models from IBM designed for low latency usage.

tools 1b 3b

15K yesterday

33 Tags
latest
d84e1e38ee39 • 822MB • yesterday
1b
d84e1e38ee39 • 822MB • yesterday
3b
157f538ae66e • 2.1GB • yesterday
1b-instruct-fp16
cee5d696574c • 2.7GB • yesterday
1b-instruct-q2_K
5965f707577c • 512MB • yesterday
1b-instruct-q3_K_L
bbb955565d3d • 711MB • yesterday
1b-instruct-q3_K_M
ea6a79b9b3cc • 659MB • yesterday
1b-instruct-q3_K_S
cab9e1d32d19 • 598MB • yesterday
1b-instruct-q4_0
9b6ee4c08df8 • 768MB • yesterday
1b-instruct-q4_1
c3aa5d722767 • 849MB • yesterday
1b-instruct-q4_K_M
d84e1e38ee39 • 822MB • yesterday
1b-instruct-q4_K_S
76e6392d08a3 • 775MB • yesterday
1b-instruct-q5_0
f49b9c68d958 • 929MB • yesterday
1b-instruct-q5_1
f1dbdb4ce40f • 1.0GB • yesterday
1b-instruct-q5_K_M
c51ea1c8bab3 • 956MB • yesterday
1b-instruct-q5_K_S
427124d0793f • 929MB • yesterday
1b-instruct-q6_K
a585cc84307b • 1.1GB • yesterday
1b-instruct-q8_0
84f7efa9cc2d • 1.4GB • yesterday
3b-instruct-fp16
9433d98bf426 • 6.8GB • yesterday
3b-instruct-q2_K
48ee4c1b7090 • 1.3GB • yesterday
3b-instruct-q3_K_L
8c76215f504a • 1.8GB • yesterday
3b-instruct-q3_K_M
cc32cde7c1fe • 1.6GB • yesterday
3b-instruct-q3_K_S
5952b23cdd5c • 1.5GB • yesterday
3b-instruct-q4_0
d144aac2b0d8 • 1.9GB • yesterday
3b-instruct-q4_1
ac91c7b62cc7 • 2.1GB • yesterday
3b-instruct-q4_K_M
157f538ae66e • 2.1GB • yesterday
3b-instruct-q4_K_S
9f5ecea51621 • 1.9GB • yesterday
3b-instruct-q5_0
e623edd5db90 • 2.3GB • yesterday
3b-instruct-q5_1
709d47478efd • 2.5GB • yesterday
3b-instruct-q5_K_M
94e89b071306 • 2.4GB • yesterday
3b-instruct-q5_K_S
c84b2a3a27fd • 2.3GB • yesterday
3b-instruct-q6_K
c196db4b29bb • 2.8GB • yesterday
3b-instruct-q8_0
f3b962d19110 • 3.6GB • yesterday