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

13.3K Pulls Updated 3 weeks ago

33 Tags

5d8ebcfcdb80 • 2.1GB • 3 weeks ago

881c120886f6 • 822MB • 3 weeks ago

5d8ebcfcdb80 • 2.1GB • 3 weeks ago

1b-instruct-fp16

dd8a59650137 • 2.7GB • 3 weeks ago

1b-instruct-q2_K

45fb71415761 • 512MB • 3 weeks ago

1b-instruct-q3_K_L

0af587389b10 • 711MB • 3 weeks ago

1b-instruct-q3_K_M

dbd52d52cd74 • 659MB • 3 weeks ago

1b-instruct-q3_K_S

9c9f36fe7340 • 598MB • 3 weeks ago

1b-instruct-q4_0

2be1f58af678 • 768MB • 3 weeks ago

1b-instruct-q4_1

6be8cc180006 • 849MB • 3 weeks ago

1b-instruct-q4_K_M

881c120886f6 • 822MB • 3 weeks ago

1b-instruct-q4_K_S

df4c220e396c • 775MB • 3 weeks ago

1b-instruct-q5_0

8799d2971267 • 929MB • 3 weeks ago

1b-instruct-q5_1

d24e3bc11356 • 1.0GB • 3 weeks ago

1b-instruct-q5_K_M

e490def0a321 • 956MB • 3 weeks ago

1b-instruct-q5_K_S

476156f71010 • 929MB • 3 weeks ago

1b-instruct-q6_K

bf15431658f2 • 1.1GB • 3 weeks ago

1b-instruct-q8_0

dcbe47c2b7c4 • 1.4GB • 3 weeks ago

3b-instruct-fp16

7ea9827d7d07 • 6.8GB • 3 weeks ago

3b-instruct-q2_K

da59c01b5ed1 • 1.3GB • 3 weeks ago

3b-instruct-q3_K_L

c8f9673aa7d5 • 1.8GB • 3 weeks ago

3b-instruct-q3_K_M

e06ddcd361b7 • 1.6GB • 3 weeks ago

3b-instruct-q3_K_S

007dc74c4762 • 1.5GB • 3 weeks ago

3b-instruct-q4_0

6daf78810309 • 1.9GB • 3 weeks ago

3b-instruct-q4_1

c688d5ecbc67 • 2.1GB • 3 weeks ago

3b-instruct-q4_K_M

5d8ebcfcdb80 • 2.1GB • 3 weeks ago

3b-instruct-q4_K_S

f8121a6f0ac4 • 1.9GB • 3 weeks ago

3b-instruct-q5_0

9522627e2632 • 2.3GB • 3 weeks ago

3b-instruct-q5_1

4c12546d1fdc • 2.5GB • 3 weeks ago

3b-instruct-q5_K_M

1c128c4b1bc6 • 2.4GB • 3 weeks ago

3b-instruct-q5_K_S

c34b92773d42 • 2.3GB • 3 weeks ago

3b-instruct-q6_K

a2a8e0a8bfa6 • 2.8GB • 3 weeks ago

3b-instruct-q8_0

3a4ef96e5b97 • 3.6GB • 3 weeks ago