LFM2.5-VL-3B
Liquid AIOpen weightsNon-reasoningReleased Aug 12, 2026
Overall score
—
Not enough evidence to rank yet
Context window
32K
32,000 tokens
Price per 1M tokens
— in · — out
No price listed
Blended price
—
3:1 input:output tokens
Output speed
—
Not measured yet
Benchmark results (10)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Multimodal
| MMMU-Pro | 30.5 | #48 of 48 |
| MMMU | 48.4 | #20 of 21 |
| SimpleVQA | 35.4 | #17 of 17 |
| RealWorldQA | 73.1 | #14 of 16 |
| CountBench | 87.3 | #7 of 8 |
| RefCOCO (avg) | 87.9 | #5 of 8 |
| OCRBench V2 | 47.5 | #5 of 5 |
Instruction following
| IFEval | 82.3 | #40 of 53 |
| IFBench | 25.8 | #42 of 42 |
Agentic & tool use
| BFCL v4 | 32.5 | #16 of 19 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price | — | — | — |
LiquidAI publishes LFM2.5-VL-3B under LFM Open License v1.0 for self-hosting and does not publish a first-party hosted token rate for this exact checkpoint. We represent the open-weight row as self-host/free-per-token before infrastructure costs.
Speed
Measured by us through OpenRouter (median of trials).
Not available on OpenRouter, so we cannot measure it yet.