Can I run it?
Can I run Kwaipilot KAT Coder V2.5 Dev on GeForce RTX 4080 Super?
Tight fit
Tight fit fully on your GeForce RTX 4080 Super at IQ3_XXS, 4K context, about 146–199 tok/s. Weights 13.8 GB plus 0.11 GB of KV cache — 14.4 GB of 14.7 GB usable.
Best quantIQ3_XXS
Working context4K
Max that fits4K
Speed146–199 tok/s
Weights13.8 GB
KV cache0.11 GB
Total14.4 GB
Usable14.7 GB
Kwaipilot KAT Coder V2.5 Dev details · GeForce RTX 4080 Super page · How this estimate is made
Run commands
llama.cpp
llama-server -hf bartowski/Kwaipilot_KAT-Coder-V2.5-Dev-GGUF:IQ3_XXS -c 4096 -ngl 40
No matching name in the Ollama library snapshot.
LM Studio
lms get bartowski/Kwaipilot_KAT-Coder-V2.5-Dev-GGUF # Then load quant IQ3_XXS
Assumptions
- Estimate, not a benchmark. Decode speed uses a memory-bandwidth model at 55–75% of peak.
- KV cache is FP16. Engine overhead assumes llama.cpp (0.3 GB + 1% of the weights). Ollama is closer to 0.9 GB + 3%, LM Studio closer to 1 GB + 3%.
- IQ3_XXS size is the real GGUF file size from Hugging Face.
- 0.5 GB is reserved because this GPU also drives a display, then 5% of the rest is left free.
- Speed is quoted at 4K context, not at the maximum context that still fits.
Larger models that also fit
- Can I run Qwen3.6 35B A3B on GeForce RTX 4080 Super? — Tight fit, UD-IQ3_S, 300–408 tok/s
- Can I run Qwen3.8 35B A3B Distill on GeForce RTX 4080 Super? — Tight fit, Q2_K, 303–413 tok/s
- Can I run Qwen3.5 35B A3B on GeForce RTX 4080 Super? — Tight fit, UD-IQ3_S, 301–411 tok/s
Updated 2026-10-24. Profile on this page is the device default plus 32 GB of DDR5-5600 system RAM.