Yes — with 3.9 GB to spare
Granite 4.1 3B at Q4_K_M fits your GeForce RTX 4060 Ti entirely on the GPU at 8K context, at an estimated 91 tokens per second. Past 57K the KV cache pushes it over — quantise the cache to q8_0, or step down a quantisation, to go longer.
Fully on GPU
8K context
Q4_K_M · 1.9 GB
Apache 2.0
Released 29 Apr 2026
Dense, small, enterprise-flavoured: tool calling and instruction following, no thinking mode.
The VRAM budget
weights 1.9 GB
Weights 1.9 GB
KV cache @ 8K 0.63 GB
Runtime overhead 0.6 GB
Free 3.9 GB of 7.0 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 6.3 GB | 7.6 GB | — | ~21 | Reference | 0.6 GB over |
| Q8_0 | 3.4 GB | 4.6 GB | 38K | 52 | −0.1% ppl | Long context |
| Q6_K | 2.6 GB | 3.8 GB | 48K | 67 | −0.4% ppl | Long context |
| Q5_K_M | 2.2 GB | 3.5 GB | 53K | 78 | −0.8% ppl | Long context |
| Q4_K_M | 1.9 GB | 3.1 GB | 57K | 91 | −1.9% ppl | Recommended |
| Q3_K_M | 1.5 GB | 2.8 GB | 62K | 113 | −5.4% ppl | Long context |
| Q2_K | 1.3 GB | 2.6 GB | 64K | 131 | −15% ppl | Long context |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.
How to run it
$ ollama pull granite4.1:3b $ OLLAMA_CONTEXT_LENGTH=8192 \ ollama run granite4.1:3b
The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.
01Download is 1.9 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02Close anything else holding VRAM. A browser with hardware acceleration can sit on 1–2 GB.
03There is room to go to 57K context on this card.