Benchmark result
LiquidAI/LFM2.5-350M-GGUF on NVIDIA GeForce RTX 4070 — 748.8 tok/s
Measured with Unsloth Studio on September 29, 2026.
What the model built
Open full screen →The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited.
How this run compares
2nd fastest of 2 runs of this model on NVIDIA GeForce RTX 4070 · median 764.4 tok/s.
sort
- 1779.9tok/s10.0Unsloth Studio · Q4_0 · KV f16
- 2748.8tok/s14.5Unsloth Studio · Q4_K_M · KV f16this run
Reproduce this run
toolUnsloth Studio
modelLiquidAI/LFM2.5-350M-GGUF (Q4_K_M)
context65,536
kv cachef16
thinkingoff
samplingtemperature 0.1 · top_p 0.1 · top_k 50 · min_p 0.15 · presence_penalty 0
clientv0.4.85+101
Set those in Unsloth Studio, then:
llm-benchmark benchmark --model "LiquidAI/LFM2.5-350M-GGUF" --tool "Unsloth Studio"The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.