Model Anatomy See what's inside open-weight AI models

Can I run FrogNano-4B-2609-GGUF on an NVIDIA B300?

Yes

Its BF16 file (8.67 GB) fits in the 259 GB an NVIDIA B300 can use, with room for ≈ 256K tokens of context.

Full analysis of FrogNano-4B-2609-GGUF Everything that runs on NVIDIA B300 Get the BF16 file on Hugging Face
ollama run hf.co/bartowski/FrogNano-4B-2609-GGUF:BF16

Files and context on NVIDIA B300

Each file of FrogNano-4B-2609-GGUF (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.

FileSizeOn NVIDIA B300 Context (16-bit)Context (8-bit)
BF168.67 GBFits256K256K
Q8_04.62 GBFits256K256K
Q6_K_L3.92 GBFits256K256K
Q6_K3.76 GBFits256K256K
Q6_K_S3.61 GBFits256K256K
Q5_K_M3.36 GBFits256K256K
Q5_K_S3.15 GBFits256K256K
Q4_K_L2.97 GBFits256K256K
Q4_12.92 GBFits256K256K
Q4_K_M2.80 GBFits256K256K
IQ4_NL2.79 GBFits256K256K
Q4_02.72 GBFits256K256K
Q4_K_S2.62 GBFits256K256K
IQ4_XS2.51 GBFits256K256K
IQ3_M2.35 GBFits256K256K
Q3_K_L2.24 GBFits256K256K
Q3_K_M2.14 GBFits256K256K
IQ3_XS2.05 GBFits256K256K
Q3_K_S2.04 GBFits256K256K
IQ3_XXS1.98 GBFits256K256K
Q2_K1.80 GBFits256K256K
IQ2_M1.65 GBFits256K256K

FrogNano-4B-2609-GGUF on other hardware

Popular models that run on NVIDIA B300

Assumes 90% of the card's memory is usable and at least 4K tokens of context for one conversation, with the context cache in 16-bit unless noted. Fits leaves at least 20% of that memory for context; Tight leaves less.