Models › huihui-ai_Mistral-Small-24B-Instruct-2501-abliterated

huihui-ai_Mistral-Small-24B-Instruct-2501-abliterated: VRAM needs and speed by GPU

At Q4_K_M, huihui-ai_Mistral-Small-24B-Instruct-2501-abliterated needs about 15.7 GB of VRAM. No measured runs yet. Speeds below are estimates. Submit a run.

23.57B parameters · license: apache-2.0 · 32,768-token context · GGUF published Feb 5, 2025 · GGUF files on Hugging Face

Memory needed per quantization

QuantFile sizeNeeds (estimate)Smallest card it fits
IQ2_XXS6.5 GB7.9 GB12 GB cards
IQ2_XS7.2 GB8.6 GB12 GB cards
IQ2_S7.5 GB8.9 GB12 GB cards
IQ2_M8.1 GB9.5 GB12 GB cards
Q2_K8.9 GB10.3 GB12 GB cards
Q2_K_L9.5 GB10.9 GB12 GB cards
IQ3_XS9.9 GB11.3 GB12 GB cards
Q3_K_S10.4 GB11.8 GB16 GB cards
IQ3_M10.7 GB12.0 GB16 GB cards
Q3_K_M11.5 GB12.9 GB16 GB cards
Q3_K_L12.4 GB13.8 GB16 GB cards
IQ4_XS12.8 GB14.1 GB16 GB cards
Q3_K_XL13.0 GB14.4 GB16 GB cards
IQ4_NL13.5 GB14.9 GB16 GB cards
Q4_013.5 GB14.9 GB16 GB cards
Q4_K_S13.5 GB14.9 GB16 GB cards
Q4_K_M14.3 GB15.7 GB24 GB cards
Q4_K_L14.8 GB16.2 GB24 GB cards
Q4_114.9 GB16.3 GB24 GB cards
Q5_K_S16.3 GB17.7 GB24 GB cards
Q5_K_M16.8 GB18.2 GB24 GB cards
Q5_K_L17.2 GB18.6 GB24 GB cards
Q6_K19.3 GB20.7 GB24 GB cards
Q6_K_L19.7 GB21.1 GB24 GB cards
Q8_025.1 GB26.4 GB32 GB cards

Expected speed on popular hardware

HardwareBest quant that fitsExpected tok/s
NVIDIA RTX 3060Q2_K34.7 estimate
NVIDIA RTX 4060 TiDoes not fit
NVIDIA RTX 3090Q6_K33.9 estimate
NVIDIA RTX 4090Q6_K37.6 estimate
NVIDIA RTX 5090Q8_045.8 estimate
AMD Radeon RX 7900 XTXQ6_K37.7 estimate
Apple M4 Pro (20-core GPU) (64 GB)Q8_09.2 estimate
Apple M3 Max (40-core GPU) (64 GB)Q8_013.5 estimate
AMD Ryzen AI Max+ 395 (Radeon 8060S)Q8_08.3 estimate

Check your own card

Too big for your card?

Run huihui-ai_Mistral-Small-24B-Instruct-2501-abliterated on a rented 24-80 GB GPU by the hour.

Rent a bigger GPU on RunPod

Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.