GGUF models by ibm-granite
48 GGUF models quantized by ibm-granite on Hugging Face. Sizes run from 164M to 33.7B parameters. The longest trained context is 1024K tokens.
GR
Granite 4.0 H Tiny Base
Granite · ibm-granite
Params6.94B
Context1024K
Q4_K_M4.2 GB
GR
Granite Speech 4.1 2b Plus
Granite · ibm-granite
Params1.63B
Context4K
Q4_K_M1.0 GB
GR
Granite 4.0 Tiny Preview
Granite · ibm-granite
Params6.67B
Context1024K
Q4_K_M4.0 GB
GR
Granite 4.0 H 350m Base
Granite · ibm-granite
Params340M
Context1024K
Q4_K_M0.2 GB
GR
Granite 4.0 Tiny Base Preview
Granite · ibm-granite
Params6.67B
Context1024K
Q4_K_M4.1 GB
GR
Granite 3.3 8b Base
Granite · ibm-granite
Params8.17B
Context128K
Q4_K_M4.9 GB
GR
Granite 3.3 2b Base
Granite · ibm-granite
Params2.53B
Context128K
Q4_K_M1.5 GB
GR
Granite 4.0 H Micro Base
Granite · ibm-granite
Params3.19B
Context1024K
Q4_K_M1.9 GB
GR
Granite 4.0 H Small Base
Granite · ibm-granite
Params32.2B
Context1024K
Q4_K_M19.5 GB
GR
granite-3b-code-instruct
Granite · ibm-granite
Params3.48B
Context2K
Q4_K_M2.1 GB
GR
granite vision 3.3 2b chart2csv preview
Granite · ibm-granite
Params2.53B
Context128K
Q4_K_M1.5 GB
GR
granite-3b-code-base
Granite · ibm-granite
Params3.48B
Context2K
Q4_K_M2.1 GB
GR
granite 34b code instruct 8k
Granite · ibm-granite
Params33.7B
Context8K
Q4_K_M21.4 GB
GR
granite 20b code instruct 8k
Granite · ibm-granite
Params20.1B
Context8K
Q4_K_M12.8 GB
GR
granite 34b code base 8k
Granite · ibm-granite
Params33.7B
Context8K
Q4_K_M21.4 GB
GR
granite 20b code base 8k
Granite · ibm-granite
Params20.1B
Context8K
Q4_K_M12.8 GB
GR
granite-8b-code-base
Granite · ibm-granite
Params8.05B
Context4K
Q4_K_M4.9 GB
GR
Granite 4.0 1b Speech
Granite · ibm-granite
Params1.84B
Context4K
Q4_K_M1.1 GB