LFM3-VL-2B on Pi 5

Snapshot 2026-10-27 · ai-tier

Raspberry Pi 5 · 16 GB RAMAll performance measurements are for this device.

Capability shows how well the model answers the benchmark questions. Speed, response time and memory are measured on Raspberry Pi 5 · 16 GB.

Capability summarises instruction following, tool calling and finding information in long prompts. Speed and memory refer to the Raspberry Pi 5. — means unavailable.
Model	Capability / 100	Output tok/s	Prompt tok/s	First token · s	Task time · s	Peak memory · MiB	Parameters · B
LFM3-VL-2BQ4_K_M · 4,096 context · reasoning off	64	10.3	56.2	7.4	30.9	2,285	1.9

Benchmark breakdown

Instruction following · IFEval	Tool calling · BFCL	Finding information · RULER
69	30	93

Test configuration

Model publisher: Liquid AI
Model parameters: 1.918 billion total
Published input support: text, image, video (metadata; not device validation)
Quantization: Q4_K_M
Context capacity: 4,096 tokens
Reasoning mode: off
CPU threads / parallel requests: 4 / 1
Temperature / seed: 0 / 42
Measurement date: 2026-10-27T18:47:39.692896+00:00
Result ID: split-6d201ec8a1d32fc1b434
Runtime revision: 19ff817fb25e9a1b838b1d7f9f4325f7bf9779c4
Artifact repository / revision: LiquidAI/LFM3-VL-2B-GGUF / 7c3b4a43c9023e80bf234fec5e826a3365c8dda8
Artifact SHA256: 56dac2e192ff0d9ab8507f4f1aeb4aed0463449bbbefaac79497f041c95f21eb
Dataset revision: a0fc00b6a07455d1efb70cc19d5ca94c5923c5261a0b41095083a41a1592a1f6
Configuration identity: e6bfc3c180e63aea31dd0a8b0c9b4e54151f8e1b2945e4cde52ab3136dbac170

Measurement limits

Speed tests read 256 input tokens and generate 128 output tokens, repeated ten times. The output speed test starts without a long conversation history. Answer time uses six fixed prompts with up to 128 output tokens, repeated three times. First-text time is measured from the application making the request. Peak memory includes the loaded model; it excludes other applications on the device. Longer prompts may need more memory. Capability uses a fixed selection of public questions, rather than the full official benchmarks.

All Pi 5 measurements Methodology Publisher model card ↗Pinned engine source ↗

Measured 2026-10-27T18:47:39.692896+00:00.