Can I run clef-flash on an RTX 4060 Ti 16 GB?
Yes, with a GGUF version
The published weights are too big, but the Q8_0 file from bartowski/Cloudflare_clef-flash-GGUF (9.68 GB) fits, with room for ≈ 139K tokens of context.
Full analysis of clef-flash
Everything that runs on RTX 4060 Ti 16 GB
Get the Q8_0 file on Hugging Face
ollama run hf.co/bartowski/Cloudflare_clef-flash-GGUF:Q8_0Files and context on RTX 4060 Ti 16 GB
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.
| File | Size | On RTX 4060 Ti 16 GB | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| As published | 18.8 GB | Doesn't fit | — | — |
| BF16 | 18.2 GB | Doesn't fit | — | — |
| Q8_0 | 9.66 GB | Fits | 140K | 256K |
| Q4_K_M | 6.49 GB | Fits | 234K | 256K |
| BF16 | 18.2 GB | Doesn't fit | — | — |
| Q8_0 | 9.68 GB | Fits | 139K | 256K |
| Q6_K_L | 8.32 GB | Fits | 180K | 256K |
| Q6_K | 7.83 GB | Fits | 194K | 256K |
| Q5_K_M | 6.98 GB | Fits | 220K | 256K |
| Q4_K_L | 6.80 GB | Fits | 225K | 256K |
| Q5_K_S | 6.66 GB | Fits | 229K | 256K |
| Q4_1 | 6.07 GB | Fits | 247K | 256K |
| Q4_K_M | 6.04 GB | Fits | 248K | 256K |
| Q4_K_S | 5.73 GB | Fits | 256K | 256K |
| IQ4_NL | 5.63 GB | Fits | 256K | 256K |
| Q4_0 | 5.60 GB | Fits | 256K | 256K |
| IQ4_XS | 5.40 GB | Fits | 256K | 256K |
| Q3_K_L | 5.24 GB | Fits | 256K | 256K |
| Q3_K_M | 5.05 GB | Fits | 256K | 256K |
| IQ3_M | 4.85 GB | Fits | 256K | 256K |
| Q3_K_S | 4.80 GB | Fits | 256K | 256K |
| Q2_K | 4.19 GB | Fits | 256K | 256K |