Skip to content

Can RTX 4080 Super run Ministral 3 14B Instruct?

Comfortable: GGUF Q5_K_M needs 10.7 GB at 4K, batch 1; headroom 5.3 GB.

Text input, FP16 KV cache, one card. Unified-memory capacity uses the GPU guide’s usable limit; negative headroom may require host offload. Speed source settings may differ from this 4K estimate.

Measured runs

No measured run for this pair. Sizing is an estimate, not a runtime test.

Quant memory at 4K

QuantTotal GBVerdictWeight sizing
GGUF Q4_K_M9.3ComfortableVerified file
GGUF Q5_K_M10.7ComfortableVerified file
GGUF Q8_015.7TightVerified file

Context capacity · GGUF Q5_K_M

ContextTotal GBVerdict
4K10.7Comfortable
16K12.8Comfortable
32K15.6Tight
128K32.1Does not fit

Generation speed

GGUF Q5_K_M: 58–77 tok/s · Derived range

Other model / GPU pairs