Card × model estimate

Qwen2.5 1.5B on RTX 4070 Ti Q4_K_M

should fit

On an RTX 4070 Ti (12 GB), Qwen2.5 1.5B in Q4_K_M needs about 2,4 Go of video memory: it should fit. Estimated generation speed: ≈ 372 tokens/s.

Not measured yet: the figures below are estimates, computed from the models already measured on this card. Measure my card

≈ 372 tok/s

Generation (estimated)

≥ 2,4 Go

Recommended VRAM (12 GB avail.)

≈ 48 W

Power (ref. model)

0,01 €/M tok

Electricity / M tokens · 0,25 €/kWh

Calculation: R1 Distill 1.5B Qwen measured at 328 tok/s on this card, 1,0 Go read per token → × 1,0 Go ÷ 0,9 Go ≈ 372 tok/s (generation is memory-bound: speed follows the amount of weights read).

What about a longer context?

ContextKV cacheEstimated VRAMFits in 12 GB?Estimated speed
4 0960,1 Go1,5 Goyes≈ 338 tok/s
8 1920,2 Go1,6 Goyes≈ 305 tok/s
16 3840,4 Go1,9 Goyes≈ 255 tok/s
32 7680,9 Go2,3 Goyes≈ 192 tok/s

Estimated VRAM = file + KV cache (full-attention layers, f16) + ~0,5 Go of buffers, excluding memory used by the system — hence the slightly higher “recommended VRAM”, which keeps that margin. Speed ≈ estimated speed × weights read ÷ (weights read + added KV cache).

Qwen2.5 1.5B on other cards

should fit
Qwen2.5 1.5B pageRTX 4070 Ti pageFeel the speed Download the GGUF (0,9 Go)

Estimates based on real measurements of other models on this card and on the model's public data (Qwen). As soon as a measurement exists, this page shows the real figures.