Model Library
Liquid AIReleased May 28, 2026LFM2.5-8B-A1B
Liquid AI · Released May 28, 2026
Intelligence score
8
At a glance
- Intelligence
- 8
- Speed
- 323.9tokens/sec
- Latency
- 6.50stime to first token
- Input price
- Not published
- Output price
- Not published
- Context window
- 33Ktokens
- Parameters
- 8.3billions
Summary
Comparison summary
LFM2.5-8B-A1B scores 8 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight models of similar size (median: 9). LFM2.5-8B-A1B generates output at 323.9 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight models of similar size (median: 101.0 t/s).
Specifications
Technical specifications
| Model name | LFM2.5-8B-A1B |
|---|---|
| Provider | Liquid AI |
| Release date | May 28, 2026 |
| Intelligence score | 8 |
| Speed | 323.9 tokens/sec |
| Latency (TTFT) | 6.50s |
| Input price | Not published |
| Output price | Not published |
| Context window | 33,000 tokens |
| Parameters | 8.3B |
| Reasoning | Yes |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | lfm 1 |
| API availability | — |
| Source | Model Intelligence legacy profile: lfm2-5-8b-a1b |
FAQ · 18
Frequently asked questions
LFM2.5-8B-A1B was released on May 28, 2026.
LFM2.5-8B-A1B was created by Liquid AI.
LFM2.5-8B-A1B scores 8 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight models of similar size (median: 9).
LFM2.5-8B-A1B generates output at 323.9 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight models of similar size (median: 101.0 t/s).
LFM2.5-8B-A1B has a time to first token (TTFT) of 6.50s (based on the median across providers serving the model), which is at the higher end compared to other open weight models of similar size (median: 2.09s).
Yes, LFM2.5-8B-A1B is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
LFM2.5-8B-A1B supports text input.
LFM2.5-8B-A1B supports text output.
No, LFM2.5-8B-A1B does not support image input. It can only process text.
No, LFM2.5-8B-A1B is not multimodal. It only supports text input.
LFM2.5-8B-A1B has a context window of 33k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, LFM2.5-8B-A1B is open weights. The model weights are publicly available and can be downloaded for self-hosting.
LFM2.5-8B-A1B has 8.3 billion parameters (1.5 billion active).
LFM2.5-8B-A1B is a Mixture of Experts (MoE) model with 8.3 billion total parameters, but only 1.5 billion active parameters are used during inference.
LFM2.5-8B-A1B is released under the lfm 1.0 license. This license allows commercial use.
LFM2.5-8B-A1B achieves a score of 8 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, LFM2.5-8B-A1B is available via API through 1 provider.
LFM2.5-8B-A1B is available through 1 API provider.