Models › AesCode-32B

AesCode-32B: VRAM needs and speed by GPU

At Q4_K_M, AesCode-32B needs about 22.1 GB of VRAM. No measured runs yet. Speeds below are estimates. Submit a run.

32.76B parameters · license: apache-2.0 · 262,144-token context · GGUF published Oct 11, 2026 · GGUF files on Hugging Face

Memory needed per quantization

QuantFile sizeNeeds (estimate)Smallest card it fits
IQ2_XXS10.6 GB12.3 GB16 GB cards
IQ2_XS11.4 GB13.0 GB16 GB cards
IQ2_S12.0 GB13.6 GB16 GB cards
IQ2_M12.7 GB14.3 GB16 GB cards
Q2_K13.0 GB14.6 GB16 GB cards
IQ3_XXS13.6 GB15.2 GB16 GB cards
IQ3_XS14.4 GB16.0 GB24 GB cards
Q3_K_S14.9 GB16.5 GB24 GB cards
IQ3_M15.5 GB17.1 GB24 GB cards
Q3_K_M16.5 GB18.1 GB24 GB cards
Q3_K_L17.5 GB19.1 GB24 GB cards
IQ4_XS18.1 GB19.7 GB24 GB cards
Q4_019.0 GB20.7 GB24 GB cards
IQ4_NL19.1 GB20.7 GB24 GB cards
Q4_K_S19.1 GB20.8 GB24 GB cards
Q4_K_M20.5 GB22.1 GB24 GB cards
Q4_120.9 GB22.5 GB24 GB cards
Q4_K_L21.1 GB22.7 GB24 GB cards
Q5_K_S22.8 GB24.4 GB32 GB cards
Q5_K_M23.8 GB25.4 GB32 GB cards
Q6_K28.0 GB29.6 GB32 GB cards
Q6_K_L28.4 GB30.0 GB32 GB cards
Q8_034.8 GB36.4 GB48 GB cards
BF1665.5 GB67.2 GB96 GB cards

Expected speed on popular hardware

HardwareBest quant that fitsExpected tok/s
NVIDIA RTX 3060Does not fit
NVIDIA RTX 4060 TiDoes not fit
NVIDIA RTX 3090Q4_K_M32.0 estimate
NVIDIA RTX 4090Q4_K_M35.5 estimate
NVIDIA RTX 5090Q6_K41.0 estimate
AMD Radeon RX 7900 XTXQ4_K_M35.6 estimate
Apple M4 Pro (20-core GPU) (64 GB)Q8_06.6 estimate
Apple M3 Max (40-core GPU) (64 GB)Q8_09.7 estimate
AMD Ryzen AI Max+ 395 (Radeon 8060S)Q8_06.0 estimate

Check your own card

Too big for your card?

Run AesCode-32B 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.