RigBench data 2026-10-22

Home › Models › Kimi K2.7 Code

Kimi K2.7 Code: VRAM needs and speed by GPU

At UD-Q4_K_XL, Kimi K2.7 Code needs about 586.7 GB of VRAM. We have 1 measured speed results for it.

Ranked #20 among open-weight models by the Artificial Analysis Intelligence Index, checked 2026-10-22T16:10:10Z. All leading open models.

1026.41B parameters (mixture-of-experts, ~32.8B active) · license: other · 262,144-token context · GGUF published 2026-06-12T18:03:48.000Z · GGUF files on Hugging Face

Measured speeds

ChipGeneration tok/sPrompt tok/sQuantBackendTestSource
Apple M3 Ultra (80-core GPU)24.6145UD-Q2_K_XLMetal tg300 (llama-bench -n 300)AI KIZAI: Mac Studio M3 Ultra 512 GB measured benchmarks details

Memory needed per quantization

QuantFile sizeNeeds (estimate)Smallest card it fits
UD-IQ1_M303.9 GB306.9 GBmulti-GPU / large unified memory
UD-IQ2_XXS317.8 GB320.8 GBmulti-GPU / large unified memory
UD-IQ2_M318.0 GB321.0 GBmulti-GPU / large unified memory
UD-Q2_K_XL339.5 GB342.5 GBmulti-GPU / large unified memory
UD-IQ3_S418.8 GB421.8 GBmulti-GPU / large unified memory
UD-Q3_K_M463.6 GB466.6 GBmulti-GPU / large unified memory
UD-Q3_K_XL463.9 GB466.9 GBmulti-GPU / large unified memory
UD-IQ4_XS495.1 GB498.1 GBmulti-GPU / large unified memory
UD-Q4_K_XL583.7 GB586.7 GBmulti-GPU / large unified memory
UD-Q8_K_XL594.5 GB597.5 GBmulti-GPU / large unified memory

Expected speed on popular hardware

HardwareBest quant that fitsExpected tok/s
NVIDIA RTX 3060Does not fit
NVIDIA RTX 4060 TiDoes not fit
NVIDIA RTX 3090Does not fit
NVIDIA RTX 4090Does not fit
NVIDIA RTX 5090Does not fit
AMD Radeon RX 7900 XTXDoes not fit
Apple M4 Pro (20-core GPU)Does not fit
Apple M3 Max (40-core GPU)Does not fit
AMD Ryzen AI Max+ 395 (Radeon 8060S)Does not fit

Check your own card

Too big for your card?

Run Kimi K2.7 Code 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.