Tier
Moderate
Headquarters
Cambridge, MA, USA
Founded
2023
Flagship
LFM
Models tracked
9
Artificial Analysis
Top intelligence
8.3
Artificial Analysis Index
Cheapest model
$0.00
blended /1M
Fastest output
426
tokens / sec
Flagship Model
Liquid AI’s top-scoring model on the Artificial Analysis Intelligence Index.
Most intelligent model
LFM2.5-8B-A1B
Released 2026-05-28
Intelligence
8.3
Coding
—
Math
—
In $/1M
$0.00
Out $/1M
$0.00
Speed t/s
340
Benchmark breakdown
GPQA Diamond
47
Humanity’s Last Exam
7
SciCode
8
τ²-Bench (agentic)
16
Terminal-Bench Hard
5
IFBench
53
Models & benchmarks
Every Liquid AI model on file — intelligence, coding, price, and speed.
| Model | Released | Intel | Coding | Math | In $/1M | Out $/1M | Speed t/s | Latency s |
|---|---|---|---|---|---|---|---|---|
| LFM2.5-8B-A1B | 2026-05-28 | 8.3 | — | — | $0.00 | $0.00 | 340 | 1.30 |
| LFM2 24B A2B | 2026-02-25 | 4.9 | — | — | $0.00 | $0.00 | 0 | 0.00 |
| LFM 40B | 2024-09-30 | 3.4 | — | — | $0.00 | $0.00 | 0 | 0.00 |
| LFM2 2.6B | 2025-09-23 | 2.7 | — | 8.3 | $0.00 | $0.00 | 0 | 0.00 |
| LFM2.5-1.2B-Instruct | 2026-01-05 | 2.7 | — | — | $0.00 | $0.00 | 0 | 0.00 |
| LFM2.5-1.2B-Thinking | 2026-01-20 | 2.7 | — | — | $0.00 | $0.00 | 0 | 0.00 |
| LFM2 8B A1B | 2025-10-07 | 1.8 | — | 25.3 | $0.00 | $0.00 | 0 | 0.00 |
| LFM2 1.2B | 2025-07-10 | 1.1 | — | 3.3 | $0.00 | $0.00 | 0 | 0.00 |
| LFM2.5-VL-1.6B | 2026-01-05 | 1.0 | — | — | $0.00 | $0.00 | 426 | 0.73 |
Trajectory
Liquid AI’s model releases over time, by intelligence.
Sep 2024 · 3.4▲ +4.9 indexMay 2026 · 8.3
House strengths
Where Liquid AI’s lineup over-indexes, averaged across its models.
Capability profile
Reasoning
23
Coding
6
Math
36
Agentic
7
Instruction
18
Best value
No pricing data to rank value.