Models8in catalog
Largest context262KLFM3-1.7B
Avg context
126K
Cheapest input
Free/M
LFM2.5-2.6B (free)
Largest output
33K
LFM2.5-2.6B (free)
Speed tiers
fast 2balanced 5deep 1
Capabilities
What Liquid models support across the catalog.
Tool use
5 / 8
63% of models
Function calling
5 / 8
63% of models
Extended thinking
6 / 8
75% of models
Streaming
8 / 8
100% of models
Prompt caching
3 / 8
38% of models
All Liquid models
Sorted by context window, largest first. Tap a row for full specs.
| Model | Context | Max output | Input $/M | Speed |
|---|---|---|---|---|
| LFM3-1.7B | 262K | 8K | — | fast |
| LFM3-1.7B (free) | 262K | 8K | — | fast |
| LFM2.5-24B-A2B | 128K | 8K | — | balanced |
| LFM2.5-2.6B | 128K | 8K | — | balanced |
| LFM2.5-2.6B (free) | 128K | 33K | Free/M | balanced |
| LFM2-24B-A2B | 33K | — | — | balanced |
| LFM2.5-1.2B-Instruct (free) | 33K | — | — | balanced |
| LFM2.5-1.2B-Thinking (free) | 33K | — | — | deep |
See rankings
How Liquid models rank against the full catalog.