← AI Hashrate

Can RTX 4080 Super 16GB run Devstral-Small-2-24B-Instruct-2512?

NVIDIA · 16 GB GDDR6X · 736 GB/s bandwidth · Mistral · 24.0B · model ctx up to 256K

No — Devstral-Small-2-24B-Instruct-2512 at Q4 needs ≈ 16.6 GB even at 4K context, beyond the RTX 4080 Super 16GB's 16 GB GDDR6X. It would only run with heavy CPU/disk offload. Decode at Q4/8K: ≈ 39.5 tok/s (estimated, batch 1). FP16 needs ≈ 53.8 GB — does not fit on this card. Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology.

Fit & speed by quant and context

QuantContextVRAM neededFits?tok/s (decode)
Q44K16.6 GBNo51.8 est.
Q48K default19.0 GBNo39.5 est.
Q432K33.4 GBNo12.8 est.
FP164K51.4 GBNo1.5 est.
FP168K default53.8 GBNo1.4 est.
FP1632K68.2 GBNo

Fits = weights + KV(ctx) + 1 GB overhead ≤ 95% of VRAM. Measured anchors are context-agnostic; the fit verdict is recomputed per context. A missing tok/s means the model is far beyond this card (offload-only territory).

VRAM breakdown at 8K context

QuantWeightsKV cacheOverheadTotal neededRTX 4080 Super 16GB VRAM
Q413.2 GB4.8 GB1 GB19.0 GB16 GB
FP1648.0 GB4.8 GB1 GB53.8 GB16 GB

Other GPUs that run Devstral-Small-2-24B-Instruct-2512

All GPUs for Devstral-Small-2-24B-Instruct-2512 →

Other models for the RTX 4080 Super 16GB

All models on the RTX 4080 Super 16GB →