On an RTX 4070 Ti (12 GB), Granite 4.1 8B in Q4_K_M needs about 6,7 Go of video memory: it should fit. Estimated generation speed: ≈ 111 tokens/s.
Not measured yet: the figures below are estimates, computed from the models already measured on this card. Measure my card
≈ 111 tok/s
Generation (estimated)
≥ 6,7 Go
Recommended VRAM (12 GB avail.)
≈ 115 W
Power (ref. model)
0,07 €/M tok
Electricity / M tokens · 0,25 €/kWh
Calculation: Gemma 4 E4B measured at 114 tok/s on this card, 5,0 Go read per token → × 5,0 Go ÷ 5,2 Go ≈ 111 tok/s (generation is memory-bound: speed follows the amount of weights read).
What about a longer context?
Context
KV cache
Estimated VRAM
Fits in 12 GB?
Estimated speed
4 096
0,6 Go
6,3 Go
yes
≈ 101 tok/s
8 192
1,3 Go
6,9 Go
yes
≈ 91,1 tok/s
16 384
2,5 Go
8,2 Go
yes
≈ 76,1 tok/s
32 768
5,0 Go
10,7 Go
yes
≈ 57,2 tok/s
65 536
10,0 Go
15,7 Go
no
—
131 072
20,0 Go
25,7 Go
no
—
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).
Estimates based on real measurements of other models on this card and on the model's public data (IBM Granite). As soon as a measurement exists, this page shows the real figures.