Mistral on NVIDIA Ada

Can I run Mixtral 8x7B on an RTX 4080 Super?

Not at Q4_K_M — it needs 28.2 GB against 14.2 GB available. You would need 2 of these cards.

Won't fit

28.2 GB of 14.2 GB · 198%
029 GB
Weights 26.4 GB
KV cache 1.0 GB
Runtime overhead 0.8 GB
Over the limit 14.0 GB

Short by 14.0 GB. You can run it with 15 of 32 layers on the RTX 4080 Super and the rest in system RAM, at roughly 8.45 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation8.45tok/s
Prompt processing1693tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Mixtral 8x7B on a RTX 4080 Super

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 87.0 GB 88.8 GB Won't fit 1.76
INT8 / W8A8 8.50 46.2 GB 48.0 GB Won't fit 3.72
Q8_0 (GGUF) 8.50 46.2 GB 48.0 GB Won't fit 3.72
FP8 (E4M3) 8.00 43.5 GB 45.3 GB Won't fit 4.08
Q6_K 6.56 35.7 GB 37.5 GB Won't fit 5.33
Q5_K_M 5.67 30.8 GB 32.6 GB Won't fit 6.38
Q5_K_S 5.52 30.0 GB 31.8 GB Won't fit 6.83
Q4_K_M 4.85 26.4 GB 28.2 GB Won't fit 8.45
Q4_K_S 4.58 24.9 GB 26.7 GB Won't fit 8.90
Q4_0 4.55 24.8 GB 26.6 GB Won't fit 9.41
AWQ 4-bit 4.25 23.5 GB 25.3 GB Won't fit 9.89
GPTQ 4-bit 4.25 23.5 GB 25.3 GB Won't fit 9.89
MXFP4 4.25 23.5 GB 25.3 GB Won't fit 9.89
IQ4_XS 4.25 23.1 GB 25.0 GB Won't fit 10.6
Q3_K_M 3.91 21.3 GB 23.1 GB Won't fit 12.0
IQ3_M 3.70 20.2 GB 22.0 GB Won't fit 13.4
IQ3_XXS 3.06 16.7 GB 18.5 GB Won't fit 20.9
Q2_K 2.63 14.4 GB 16.2 GB Won't fit 34.6
IQ2_XXS 2.06 11.3 GB 13.1 GB Fits, but tight 17K 97.8
IQ1_M 1.75 9.6 GB 11.4 GB Runs comfortably 30K 111

Also worth checking