Liquid AILanguage modelOpen weights
LFM3-1.7B-Base
Summary
LFM3-1.7B-Base is the pre-trained text checkpoint with 1.73 billion parameters, a 128,000-token vocabulary, and 262,144 context tokens. Liquid AI reports 22 trillion total pre-training tokens. The model is intended for substantial fine-tuning.
The base model card contains no base-specific performance benchmarks. Results for post-trained LFM3-1.7B therefore cannot be transferred to this checkpoint. The weights use the LFM Open License v1.0.
All specifications
Architecture
- Architecture
- Dense tri-hybrid architecture with 24 layers: 13 gated short convolutions, 5 sliding-window attention layers, and 6 global attention layers.
- Source: Architecture
Input and output
Licence and access
Further details
- Pre-training
- The model card reports 22 trillion pre-training tokens, including 3.5 trillion tokens of curated mid-training.
- Source: Pre-training
- Vocabulary and languages
- 128,000 tokens and 16 listed languages. The languages match the LFM3 coverage listed on the model card.
- Source: Vocabulary and languages
Frequently asked questions about LFM3-1.7B-Base
How large is the context window of LFM3-1.7B-Base?
262,144 tokens. That is how much text LFM3-1.7B-Base reads at once, about 390 A4 pages.
Page count as a rule of thumb from Artificial Analysis
Can I run LFM3-1.7B-Base locally?
Yes, the weights are open.
Stated by Liquid AI
- Open weights
- yes
- License
- LFM Open License v1.0. See the license text for its terms.
- Parameters
- 1.7B
- active per token
- n/a
Sources and data date
| Type | Evidence and data date |
|---|---|
| Input | Documented modalities. Retrieved October 29, 2026. |
| Output | Documented modalities. Retrieved October 29, 2026. |
| Research status | Research date October 29, 2026. 61 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not published | Base-checkpoint benchmarks Note The reviewed LFM3-1.7B-Base model card publishes no base-specific performance benchmarks. Results for post-trained LFM3-1.7B are not transferred. |
| Not found in the inspected sources | Independent evaluation Note The reviewed manufacturer sources do not link an independent benchmark evaluation of the base checkpoint. |
| Not found in the inspected sources | Technical paper or report Note The model card cites the LFM3 announcement but does not link a separate technical report or research paper for this checkpoint. |
| Not found in the inspected sources | Vals rows in captured payloads Note Nine complete Vals payloads dated 28 October were checked; no exact model ID or unambiguous model label. Limited to these saved benchmarks. |
| Unresolved | Uncaptured Arena Code/Speech leaderboards Note Text, Agent including Code/Chat/Work, and twelve WebDev views were captured. A separate general Code and Arena Speech leaderboard were not saved as complete tables. |
| Unresolved | Artificial Analysis catalog reconciliation Note The full current AA model and benchmark catalog was not re-reconciled for this profile in this sidecar. This sidecar makes no AA absence claim; native AA imports remain authoritative. |
| Additional source | Liquid AI LFM3 release announcement (retrieved October 29, 2026) · Liquid AI LFM3 release announcement · Editorial description reviewed October 29, 2026 |
| Additional source | Liquid AI model documentation (retrieved October 29, 2026) |
| Additional source | Liquid AI LFM3 Hugging Face collection (retrieved October 29, 2026) · LFM3 collection · Editorial description reviewed October 29, 2026 |
| Additional source | Hugging Face model metadata (retrieved October 29, 2026) |
| Additional source | Model metadata (retrieved October 29, 2026)123 |