Yes — with 12.2 GB to spare
LFM2.5 2.6B at Q4_K_M fits your Radeon RX 7900 GRE entirely on the GPU at 8K context, at an estimated 230 tokens per second. There is room for its full 128K window.
Fully on GPU
8K context
Q4_K_M · 1.5 GB
LFM Open License v1.0
Released 28 Jul 2026
New this month
Not in the Ollama library
Convolution-heavy hybrid for CPUs and NPUs: 22 of 30 blocks keep no KV cache at all.
The VRAM budget
weights 1.5 GB
Weights 1.5 GB
KV cache @ 8K 0.13 GB
Runtime overhead 0.6 GB
Free 12.2 GB of 14.4 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 5.0 GB | 5.8 GB | 128K | 69 | Reference | Long context |
| Q8_0 | 2.7 GB | 3.4 GB | 128K | 131 | −0.1% ppl | Long context |
| Q6_K | 2.1 GB | 2.8 GB | 128K | 169 | −0.4% ppl | Long context |
| Q5_K_M | 1.8 GB | 2.5 GB | 128K | 196 | −0.8% ppl | Long context |
| Q4_K_M | 1.5 GB | 2.2 GB | 128K | 230 | −1.9% ppl | Recommended |
| Q3_K_M | 1.2 GB | 2.0 GB | 128K | 284 | −5.4% ppl | Long context |
| Q2_K | 1.1 GB | 1.8 GB | 128K | 331 | −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 30 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
$ llama-server \
-hf LiquidAI/LFM2.5-2.6B:Q4_K_M \
-c 8192 -ngl 99
The engine underneath most of the others. Every knob is exposed. More on llama.cpp.
01Download is 1.5 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 the model's full 128K context on this card.