Card × model estimate

Phi-3.5 mini 3.8B on RTX 4070 Ti Q4_K_M

should fit

On an RTX 4070 Ti (12 GB), Phi-3.5 mini 3.8B in Q4_K_M needs about 3,7 Go of video memory: it should fit. Estimated generation speed: ≈ 190 tokens/s.

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

≈ 190 tok/s

Generation (estimated)

≥ 3,7 Go

Recommended VRAM (12 GB avail.)

≈ 106 W

Power (ref. model)

0,04 €/M tok

Electricity / M tokens · 0,25 €/kWh

Calculation: Gemma 3n E2B measured at 163 tok/s on this card, 2,6 Go read per token → × 2,6 Go ÷ 2,2 Go ≈ 190 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 0961,5 Go4,2 Goyes≈ 121 tok/s
8 1923,0 Go5,7 Goyes≈ 84,6 tok/s
16 3846,0 Go8,7 Goyes≈ 52,8 tok/s
32 76812,0 Go14,7 Gono—
65 53624,0 Go26,7 Gono—
131 07248,0 Go50,7 Gono—

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).

Phi-3.5 mini 3.8B on other cards

should fit
Phi-3.5 mini 3.8B pageRTX 4070 Ti pageFeel the speed Download the GGUF (2,2 Go)

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