Mistral on NVIDIA Ampere

Can I run Mistral 7B v0.3 on an RTX A6000?

Yes. Mistral 7B v0.3 at Q4_K_M uses 5.9 GB of the 44.3 GB available on a RTX A6000, and runs at about 102 tokens per second. You can push the context to 32K.

Runs comfortably

5.9 GB of 44.3 GB · 13%
044 GB
Weights 4.1 GB
KV cache 1.0 GB
Runtime overhead 0.8 GB

Mistral 7B v0.3 at Q4_K_M leaves 38.4 GB spare on a RTX A6000. There is room to raise the context length or move up a quantisation level.

Generation102tok/s
Prompt processing2172tok/s
Max context32Ktokens
KV per 1K tokens0GB

Every quantisation of Mistral 7B v0.3 on a RTX A6000

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 13.5 GB 15.3 GB Runs comfortably 32K 34.1
INT8 / W8A8 8.50 7.1 GB 8.9 GB Runs comfortably 32K 62.4
Q8_0 (GGUF) 8.50 7.1 GB 8.9 GB Runs comfortably 32K 62.4
FP8 (E4M3) 8.00 6.8 GB 8.6 GB Runs comfortably 32K 65.5
Q6_K 6.56 5.5 GB 7.4 GB Runs comfortably 32K 78.4
Q5_K_M 5.67 4.8 GB 6.6 GB Runs comfortably 32K 89.4
Q5_K_S 5.52 4.7 GB 6.5 GB Runs comfortably 32K 91.5
Q4_K_M 4.85 4.1 GB 5.9 GB Runs comfortably 32K 102
AWQ 4-bit 4.25 4.0 GB 5.8 GB Runs comfortably 32K 106
GPTQ 4-bit 4.25 4.0 GB 5.8 GB Runs comfortably 32K 106
MXFP4 4.25 4.0 GB 5.8 GB Runs comfortably 32K 106
Q4_K_S 4.58 3.9 GB 5.7 GB Runs comfortably 32K 107
Q4_0 4.55 3.9 GB 5.7 GB Runs comfortably 32K 108
IQ4_XS 4.25 3.6 GB 5.4 GB Runs comfortably 32K 114
Q3_K_M 3.91 3.3 GB 5.2 GB Runs comfortably 32K 122
IQ3_M 3.70 3.2 GB 5.0 GB Runs comfortably 32K 127
IQ3_XXS 3.06 2.7 GB 4.5 GB Runs comfortably 32K 148
Q2_K 2.63 2.3 GB 4.1 GB Runs comfortably 32K 166
IQ2_XXS 2.06 1.8 GB 3.7 GB Runs comfortably 32K 197
IQ1_M 1.75 1.6 GB 3.4 GB Runs comfortably 32K 220

Also worth checking

Mistral 7B v0.3 on every GPU
Full hardware tableEverything that runs on a RTX A6000
Full model tableWhy the KV cache is the part that bites
Guide