Models › AesCode-8B

AesCode-8B: VRAM needs and speed by GPU

At Q4_K_M, AesCode-8B needs about 6.2 GB of VRAM. No measured runs yet. Speeds below are estimates. Submit a run.

8.19B 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_M3.1 GB4.1 GB8 GB cards
Q2_K3.3 GB4.3 GB8 GB cards
IQ3_XXS3.7 GB4.7 GB8 GB cards
Q3_K_S3.9 GB4.9 GB8 GB cards
IQ3_XS3.9 GB4.9 GB8 GB cards
Q3_K_M4.0 GB5.0 GB8 GB cards
Q3_K_L4.2 GB5.2 GB8 GB cards
IQ3_M4.5 GB5.5 GB8 GB cards
IQ4_XS4.7 GB5.7 GB8 GB cards
Q4_04.9 GB5.9 GB8 GB cards
Q4_K_S4.9 GB5.9 GB8 GB cards
IQ4_NL5.2 GB6.2 GB8 GB cards
Q4_K_M5.2 GB6.2 GB8 GB cards
Q4_15.3 GB6.3 GB8 GB cards
Q4_K_L5.6 GB6.6 GB8 GB cards
Q5_K_S5.9 GB6.9 GB8 GB cards
Q5_K_M6.2 GB7.2 GB8 GB cards
Q6_K_S6.9 GB7.9 GB12 GB cards
Q6_K7.1 GB8.1 GB12 GB cards
Q6_K_L7.5 GB8.5 GB12 GB cards
Q8_08.7 GB9.7 GB12 GB cards
BF1616.4 GB17.4 GB24 GB cards

Expected speed on popular hardware

HardwareBest quant that fitsExpected tok/s
NVIDIA RTX 3060Q8_035.4 estimate
NVIDIA RTX 4060 TiQ5_K_M39.2 estimate
NVIDIA RTX 3090Q8_075.3 estimate
NVIDIA RTX 4090Q8_083.4 estimate
NVIDIA RTX 5090Q8_0132 estimate
AMD Radeon RX 7900 XTXQ8_083.8 estimate
Apple M4 Pro (20-core GPU) (64 GB)Q8_026.6 estimate
Apple M3 Max (40-core GPU) (64 GB)Q8_038.8 estimate
AMD Ryzen AI Max+ 395 (Radeon 8060S)Q8_023.9 estimate

Check your own card

Too big for your card?

Run AesCode-8B 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.