Can I run it?
Can I run Swift Qwen3.8 27B on GeForce RTX 3060 Ti?
Partial offload
Partial offload on your GeForce RTX 3060 Ti: Q4_K_M at 4K context, 24 of 64 layers on the GPU and 10.4 GB in system RAM. Expect about 3.9–5.3 tok/s.
Best quantQ4_K_M
Working context4K
Max that fits4K
Speed3.9–5.3 tok/sEstimate from memory bandwidth, not a benchmark. Method.
Weights16.8 GB
KV cache0.32 GB
Total17.6 GB
Usable7.13 GB
Swift Qwen3.8 27B details · GeForce RTX 3060 Ti page · How this estimate is made
Run commands
llama.cpp
llama-server -hf ukisai/Swift-Qwen3.8-27B-GGUF:Q4_K_M -c 4096 -ngl 24
No matching name in the Ollama library snapshot.
LM Studio
lms get ukisai/Swift-Qwen3.8-27B-GGUF # Then load quant Q4_K_M
Assumptions
- Estimate from memory bandwidth, not a benchmark. Decode speed uses 55–75% of peak bandwidth.
- 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%.
- Q4_K_M 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.
- About 24 of 64 layers stay on the GPU. 10.4 GB is read from system RAM.
Models that fit this machine more comfortably
- Can I run Qwen3.8 27B on GeForce RTX 3060 Ti? — Heavily degraded, UD-IQ1_M, 35–48 tok/s (estimate)
- gemma 3 27b it — Heavily degraded, UD-IQ1_S, 35–48 tok/s (estimate)
- medgemma 27b text it — Heavily degraded, UD-IQ1_S, 33–45 tok/s (estimate)
Updated 2026-10-24. Profile on this page is the device default plus 32 GB of DDR5-5600 system RAM.