Yes — with 12.9 GB to spare
LFM2.5 8B-A1B at Q4_K_M fits your RTX 4000 Ada entirely on the GPU at 8K context, at an estimated 99 tokens per second. There is room for its full 125K window.
Fully on GPU
8K context
Q4_K_M · 4.8 GB
LFM Open License v1.0
Released 28 May 2026
An 8B MoE with ~1.5B active, aimed at laptops without a GPU. Licence is permissive below $10M revenue.
The VRAM budget
weights 4.8 GB
Weights 4.8 GB
KV cache @ 8K 0.09 GB
Runtime overhead 0.6 GB
Free 12.9 GB of 18.4 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 15.8 GB | 16.5 GB | 125K | 30 | Reference | Long context |
| Q8_0 | 8.4 GB | 9.1 GB | 125K | 56 | −0.1% ppl | Long context |
| Q6_K | 6.5 GB | 7.2 GB | 125K | 73 | −0.4% ppl | Long context |
| Q5_K_M | 5.6 GB | 6.3 GB | 125K | 85 | −0.8% ppl | Long context |
| Q4_K_M | 4.8 GB | 5.5 GB | 125K | 99 | −1.9% ppl | Recommended |
| Q3_K_M | 3.9 GB | 4.5 GB | 125K | 123 | −5.4% ppl | Long context |
| Q2_K | 3.3 GB | 4.0 GB | 125K | 143 | −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 6 of its 24 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 lfm2.5:8b $ OLLAMA_CONTEXT_LENGTH=8192 \ ollama run lfm2.5:8b
The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.
01Download is 4.8 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 125K context on this card.