LFM2.5-24B-A2B
-oOdUn61bDRSykk4H951SoDOHzqoOe1.png)
- Context
- 128K
- Listed
- 2d ago
- Parameters
- 24B
- Input / 1M tokens
- $0.05
- Output / 1M tokens
- $0.70
- Cached input / 1M tokens
- $0.02
Sources:OpenRouter
Benchmarks
DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.
Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.
No benchmark results have been matched to this model yet.
Specifications
- Developer
- Liquid AI
- Model family
- —
- Type
- Language model
- Predecessor
- —
- Successor
- —
- Listed on OpenRouter
- October 13, 2026
- Model ID
liquid/lfm-2.5-24b-a2b- Openness
- Open weights
- License
- —
- Parameters
- 24B
- Input modalities
- text
- Output modalities
- text
- Modality
- text->text
- Native context
- 125K
- Tokenizer
- Other
- Max output tokens
- 8K
- Hugging Face
- LiquidAI/LFM2.5-24B-A2B
- Source freshness
- Endpoints synced Oct 15, 2026, 12:18 PM UTC
Open-weight builds
Ollama variants
Every published Ollama tag matched to this model, including parameter editions, instruction/base variants, and quantized builds.
| Variant | Parameters | Type | Quantization | Size | Context | Input |
|---|---|---|---|---|---|---|
| lfm2.5:8b-a1b-bf16 | 8B | — | BF16 | 17 GB | 125K | text |
| lfm2.5:latest | 8B | — | Q4_K_M | 5.2 GB | 125K | text |
| lfm2.5:8b-a1b-q8_0 | 8B | — | Q8_0 | 9 GB | 125K | text |
Infrastructure
Provider availability
| Provider | Endpoint | Context | Input /M tokens | Output /M tokens | Cached input /M tokens | Cache write /M tokens | Uptime 1d | Data handling |
|---|---|---|---|---|---|---|---|---|
| liquid/fp8 | 128K | $0.05 | $0.70 | $0.02 | — | 100.00% | ZDR: YesTraining: Unknown Prompt logging: Unknown Retention: Unknown Checked 2026-10-15 |
Provider data:OpenRouter
Deployment graph
Can I run this?
Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.