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

Can I run clef-flash on an RTX 5090?

Yes

Its published weights (18.8 GB) fit in the 28.8 GB an RTX 5090 can use, with room for ≈ 256K tokens of context.

Full analysis of clef-flash Everything that runs on RTX 5090 Get the weights on Hugging Face
hf download Cloudflare/clef-flash

Files and context on RTX 5090

Each file of clef-flash (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 RTX 5090 Context (16-bit)Context (8-bit)
As published18.8 GBFits256K256K
BF16ggml-org/Clef-Flash-GGUF18.2 GBFits256K256K
Q8_0ggml-org/Clef-Flash-GGUF9.66 GBFits256K256K
Q4_K_Mggml-org/Clef-Flash-GGUF6.49 GBFits256K256K
BF16bartowski/Cloudflare_clef-flash-GGUF18.2 GBFits256K256K
Q8_0bartowski/Cloudflare_clef-flash-GGUF9.68 GBFits256K256K
Q6_K_Lbartowski/Cloudflare_clef-flash-GGUF8.32 GBFits256K256K
Q6_Kbartowski/Cloudflare_clef-flash-GGUF7.83 GBFits256K256K
Q5_K_Mbartowski/Cloudflare_clef-flash-GGUF6.98 GBFits256K256K
Q4_K_Lbartowski/Cloudflare_clef-flash-GGUF6.80 GBFits256K256K
Q5_K_Sbartowski/Cloudflare_clef-flash-GGUF6.66 GBFits256K256K
Q4_1bartowski/Cloudflare_clef-flash-GGUF6.07 GBFits256K256K
Q4_K_Mbartowski/Cloudflare_clef-flash-GGUF6.04 GBFits256K256K
Q4_K_Sbartowski/Cloudflare_clef-flash-GGUF5.73 GBFits256K256K
IQ4_NLbartowski/Cloudflare_clef-flash-GGUF5.63 GBFits256K256K
Q4_0bartowski/Cloudflare_clef-flash-GGUF5.60 GBFits256K256K
IQ4_XSbartowski/Cloudflare_clef-flash-GGUF5.40 GBFits256K256K
Q3_K_Lbartowski/Cloudflare_clef-flash-GGUF5.24 GBFits256K256K
Q3_K_Mbartowski/Cloudflare_clef-flash-GGUF5.05 GBFits256K256K
IQ3_Mbartowski/Cloudflare_clef-flash-GGUF4.85 GBFits256K256K
Q3_K_Sbartowski/Cloudflare_clef-flash-GGUF4.80 GBFits256K256K
Q2_Kbartowski/Cloudflare_clef-flash-GGUF4.19 GBFits256K256K

clef-flash on other hardware

Popular models that run on RTX 5090

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.