Yes — with 3.4 GB to spare
Qwen3.5 9B at Q4_K_M fits your GeForce RTX 2080 Ti entirely on the GPU at 8K context, at an estimated 69 tokens per second. Past 117K 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 · 5.4 GB
Apache 2.0
Released 28 Feb 2026
Vision
The default for 8–12 GB cards in 2026: beats every older 8B on every published benchmark, with vision.
The VRAM budget
weights 5.4 GB
Weights 5.4 GB
KV cache @ 8K 0.25 GB
Runtime overhead 0.6 GB
Free 3.4 GB of 9.7 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 18.0 GB | 18.8 GB | — | ~3.6 | Reference | 9.1 GB over |
| Q8_0 | 9.5 GB | 10.4 GB | — | ~23 | −0.1% ppl | 0.7 GB over |
| Q6_K | 7.4 GB | 8.2 GB | 55K | 51 | −0.4% ppl | Long context |
| Q5_K_M | 6.4 GB | 7.2 GB | 87K | 59 | −0.8% ppl | Long context |
| Q4_K_M | 5.4 GB | 6.3 GB | 117K | 69 | −1.9% ppl | Recommended |
| Q3_K_M | 4.4 GB | 5.2 GB | 150K | 85 | −5.4% ppl | Long context |
| Q2_K | 3.8 GB | 4.6 GB | 170K | 99 | −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. Only 8 of its 32 blocks keep a per-token KV cache; the rest are linear-attention, Mamba or convolution blocks with a fixed-size state.
How to run it
$ ollama pull qwen3.5:9b $ OLLAMA_CONTEXT_LENGTH=8192 \ ollama run qwen3.5:9b
The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.
01Download is 5.4 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 117K context on this card.