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

Can I run clef-flash on an RTX 5060?

Yes, with a GGUF version

The published weights are too big, but the Q5_K_M file from bartowski/Cloudflare_clef-flash-GGUF (6.98 GB) fits, with room for ≈ 5K tokens of context.

Full analysis of clef-flash Everything that runs on RTX 5060 Get the Q5_K_M file on Hugging Face
ollama run hf.co/bartowski/Cloudflare_clef-flash-GGUF:Q5_K_M

Files and context on RTX 5060

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 5060 Context (16-bit)Context (8-bit)
As published18.8 GBDoesn't fit——
BF16ggml-org/Clef-Flash-GGUF18.2 GBDoesn't fit——
Q8_0ggml-org/Clef-Flash-GGUF9.66 GBDoesn't fit——
Q4_K_Mggml-org/Clef-Flash-GGUF6.49 GBTight20K39K
BF16bartowski/Cloudflare_clef-flash-GGUF18.2 GBDoesn't fit——
Q8_0bartowski/Cloudflare_clef-flash-GGUF9.68 GBDoesn't fit——
Q6_K_Lbartowski/Cloudflare_clef-flash-GGUF8.32 GBDoesn't fit——
Q6_Kbartowski/Cloudflare_clef-flash-GGUF7.83 GBDoesn't fit——
Q5_K_Mbartowski/Cloudflare_clef-flash-GGUF6.98 GBTight5K10K
Q4_K_Lbartowski/Cloudflare_clef-flash-GGUF6.80 GBTight11K21K
Q5_K_Sbartowski/Cloudflare_clef-flash-GGUF6.66 GBTight15K29K
Q4_1bartowski/Cloudflare_clef-flash-GGUF6.07 GBTight32K64K
Q4_K_Mbartowski/Cloudflare_clef-flash-GGUF6.04 GBTight33K66K
Q4_K_Sbartowski/Cloudflare_clef-flash-GGUF5.73 GBFits42K85K
IQ4_NLbartowski/Cloudflare_clef-flash-GGUF5.63 GBFits45K90K
Q4_0bartowski/Cloudflare_clef-flash-GGUF5.60 GBFits46K92K
IQ4_XSbartowski/Cloudflare_clef-flash-GGUF5.40 GBFits52K104K
Q3_K_Lbartowski/Cloudflare_clef-flash-GGUF5.24 GBFits57K114K
Q3_K_Mbartowski/Cloudflare_clef-flash-GGUF5.05 GBFits63K125K
IQ3_Mbartowski/Cloudflare_clef-flash-GGUF4.85 GBFits68K137K
Q3_K_Sbartowski/Cloudflare_clef-flash-GGUF4.80 GBFits70K140K
Q2_Kbartowski/Cloudflare_clef-flash-GGUF4.19 GBFits88K176K

clef-flash on other hardware

Popular models that run on RTX 5060

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.