LFM2.5-2.6B
Summary
LFM2.5-2.6B is an open-weight language model from Liquid, released on 11 Aug 2026. It accepts text and generates text, with a 128K-token context window and up to 8K output tokens per response. It is scheduled for retirement on 13 Nov 2026.
LFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or...
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Liquid |
|---|---|
| Release date | 11 Aug 2026 [OpenRouter] |
| Status | Generally available |
| Weights | Open weights [Hugging Face] |
| License | lfm1.0 [Hugging Face] |
| Parameters | 2.7B [Hugging Face] |
| Context window | 128K (128,000 tokens) [OpenRouter] |
| Max output | 8K tokens [OpenRouter] |
| Input | Text [OpenRouter] |
| Output | Text [OpenRouter] |
| Reasoning mode | Yes [OpenRouter] |
| Tool calling | Yes [OpenRouter] |
| Structured output | Yes [OpenRouter] |
| Retirement date | 13 Nov 2026 [OpenRouter] |
| Hugging Face | LiquidAI/LFM2.5-2.6B [Hugging Face] |
LFM2.5-2.6B API pricing by provider
USD per million tokens. Sorted by input price.
We do not currently track a paid API listing for this model.
Running LFM2.5-2.6B locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 1.6 GB | 2.6 GB | 3.0 GB |
| Q8_0 (8-bit) | 2.9 GB | 3.8 GB | 4.2 GB |
| FP16 / BF16 (16-bit) | 5.4 GB | 6.3 GB | 6.7 GB |
At 4-bit with an 8K conversation, LFM2.5-2.6B needs about 2.6 GB, so it fits on a graphics card with 8 GB or more, such as the NVIDIA GeForce RTX 4060 8 GB. Check your own hardware, or see Running local LLMs on an 8 GB GPU on our sister site AIPCs.
Frequently asked questions
What is the context window of LFM2.5-2.6B?
LFM2.5-2.6B has a context window of 128,000 tokens (128K) and can generate up to 8,192 tokens in a single response, according to OpenRouter.
Is LFM2.5-2.6B open source?
LFM2.5-2.6B is an open-weight model: its weights are published on Hugging Face as LiquidAI/LFM2.5-2.6B under the lfm1.0 license. Check the license terms before commercial use.
When was LFM2.5-2.6B released?
Liquid released LFM2.5-2.6B on 11 Aug 2026, according to OpenRouter.
How much VRAM does LFM2.5-2.6B need?
About 2.6 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 3.8 GB at 8-bit. The 4-bit model fits on a graphics card with 8 GB or more, such as the NVIDIA GeForce RTX 4060 8 GB. These are estimates from the model's published parameter count and architecture.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Muse Glimmer 30B | Meta | 10 Aug 2026 | 131K | $0.35 | $1.50 | Open |
| Granite 4.2 8B | IBM | 25 Aug 2026 | 131K | $0.06 | $0.15 | Open |
| Granite 4.2 8B | IBM | 31 Aug 2026 | 131K | – | – | Open |
| MiniCPM5-2B | Openbmb | 6 Sept 2026 | 131K | – | – | Open |
| Aion-3.0 | Aion Labs | 7 Jul 2026 | 131K | – | – | – |
| Aion-3.0-Mini | Aion Labs | 7 Jul 2026 | 131K | – | – | – |
Change history
- No longer listed on openrouter
- Retirement date changed from none to 13 Nov 2026
- License changed from none to lfm1.0
- Parameter count changed from none to 2.7B
- Added to the directory
- Now listed on openrouter