Models › basal-omni-1.0-ultra

basal-omni-1.0-ultra: VRAM needs and speed by GPU

At Q4_K_M, basal-omni-1.0-ultra needs about 21.1 GB of VRAM. No measured runs yet. Speeds below are estimates. Submit a run.

30.7B parameters · license: apache-2.0 · 262,144-token context · GGUF published Oct 18, 2026 · GGUF files on Hugging Face

Memory needed per quantization

QuantFile sizeNeeds (estimate)Smallest card it fits
IQ2_XXS9.2 GB10.7 GB12 GB cards
IQ2_XS9.6 GB11.2 GB12 GB cards
IQ2_S10.2 GB11.8 GB16 GB cards
IQ2_M11.3 GB12.9 GB16 GB cards
Q2_K11.9 GB13.5 GB16 GB cards
IQ3_XXS13.3 GB14.9 GB16 GB cards
Q3_K_S14.0 GB15.5 GB24 GB cards
IQ3_XS14.0 GB15.6 GB24 GB cards
Q3_K_M14.7 GB16.3 GB24 GB cards
Q3_K_L15.6 GB17.2 GB24 GB cards
IQ3_M16.4 GB18.0 GB24 GB cards
IQ4_XS17.2 GB18.8 GB24 GB cards
Q4_018.1 GB19.6 GB24 GB cards
Q4_K_S18.3 GB19.9 GB24 GB cards
IQ4_NL19.4 GB21.0 GB24 GB cards
Q4_K_M19.5 GB21.1 GB24 GB cards
Q4_119.8 GB21.3 GB24 GB cards
Q4_K_L20.9 GB22.5 GB24 GB cards
Q5_K_S22.0 GB23.6 GB32 GB cards
Q5_K_M23.5 GB25.1 GB32 GB cards
Q6_K_S25.8 GB27.3 GB32 GB cards
Q6_K27.0 GB28.5 GB32 GB cards
Q6_K_L28.3 GB29.9 GB32 GB cards
Q8_032.6 GB34.2 GB48 GB cards
BF1661.4 GB63.0 GB96 GB cards

Expected speed on popular hardware

HardwareBest quant that fitsExpected tok/s
NVIDIA RTX 3060IQ2_XXS33.6 estimate
NVIDIA RTX 4060 TiDoes not fit
NVIDIA RTX 3090Q4_K_M33.6 estimate
NVIDIA RTX 4090Q4_K_M37.2 estimate
NVIDIA RTX 5090Q6_K42.6 estimate
AMD Radeon RX 7900 XTXQ4_K_M37.4 estimate
Apple M4 Pro (20-core GPU) (64 GB)Q8_07.1 estimate
Apple M3 Max (40-core GPU) (64 GB)Q8_010.4 estimate
AMD Ryzen AI Max+ 395 (Radeon 8060S)Q8_06.4 estimate

Check your own card

Too big for your card?

Run basal-omni-1.0-ultra 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.