Models › HybridAgent-14B-Agentic

HybridAgent-14B-Agentic: VRAM needs and speed by GPU

At Q4_K_M, HybridAgent-14B-Agentic needs about 10.5 GB of VRAM. No measured runs yet. Speeds below are estimates. Submit a run.

14.77B parameters · license: mit · 32,768-token context · GGUF published Oct 18, 2026 · GGUF files on Hugging Face

Memory needed per quantization

QuantFile sizeNeeds (estimate)Smallest card it fits
IQ2_S5.1 GB6.3 GB8 GB cards
IQ2_M5.6 GB6.7 GB8 GB cards
Q2_K5.8 GB7.0 GB8 GB cards
IQ3_XXS6.6 GB7.8 GB12 GB cards
Q3_K_S6.9 GB8.0 GB12 GB cards
IQ3_XS6.9 GB8.1 GB12 GB cards
Q3_K_M7.2 GB8.3 GB12 GB cards
Q3_K_L7.6 GB8.7 GB12 GB cards
IQ3_M8.0 GB9.2 GB12 GB cards
IQ4_XS8.4 GB9.5 GB12 GB cards
Q4_08.7 GB9.8 GB12 GB cards
Q4_K_S8.9 GB10.0 GB12 GB cards
IQ4_NL9.3 GB10.4 GB12 GB cards
Q4_K_M9.3 GB10.5 GB12 GB cards
Q4_19.5 GB10.7 GB12 GB cards
Q4_K_L10.0 GB11.2 GB12 GB cards
Q5_K_S10.6 GB11.8 GB16 GB cards
Q5_K_M11.2 GB12.4 GB16 GB cards
Q6_K_S12.4 GB13.5 GB16 GB cards
Q6_K12.9 GB14.1 GB16 GB cards
Q6_K_L13.5 GB14.7 GB16 GB cards
Q8_015.7 GB16.9 GB24 GB cards
BF1629.5 GB30.7 GB32 GB cards

Expected speed on popular hardware

HardwareBest quant that fitsExpected tok/s
NVIDIA RTX 3060Q4_K_M33.1 estimate
NVIDIA RTX 4060 TiQ2_K42.3 estimate
NVIDIA RTX 3090Q8_041.8 estimate
NVIDIA RTX 4090Q8_046.3 estimate
NVIDIA RTX 5090Q8_073.2 estimate
AMD Radeon RX 7900 XTXQ8_046.5 estimate
Apple M4 Pro (20-core GPU) (64 GB)Q8_014.7 estimate
Apple M3 Max (40-core GPU) (64 GB)Q8_021.5 estimate
AMD Ryzen AI Max+ 395 (Radeon 8060S)Q8_013.2 estimate

Check your own card

Too big for your card?

Run HybridAgent-14B-Agentic 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.