Compare / Spark-X2.6 1.7B vs LFM2.5 1.2B Thinking
Spark-X2.6 1.7B vs LFM2.5 1.2B Thinking
Updated November 2026
Verdict
Spark-X2.6 1.7B scores higher on the Small LM Index: 50.1 vs 30.1. On the 5 benchmarks both labs report, Spark-X2.6 1.7B wins 4 and LFM2.5 1.2B Thinking wins 1. Both need about 1.7 GB of memory at 4-bit quantization. Neither model has a hosted API on OpenRouter, so you would run both yourself.
Pick Spark-X2.6 1.7B if…
- you want the strongest model overall
- you need math and reasoning
- you need precise instruction following
- you work with long documents
- you need a permissive license (Apache-2.0)
Pick LFM2.5 1.2B Thinking if…
- you build agents that call tools
Specifications
| Spark-X2.6 1.7B | LFM2.5 1.2B Thinking | |
|---|---|---|
| Small LM Index | 50.1 (likely 47–53) | 30.1 (likely 27–34) |
| Lab | SparkLLM (XHToken) · China | Liquid AI · United States |
| Released | 2026-10-19 | 2026-01-20 |
| Parameters | 1.7B (dense) | 1.2B (dense) |
| Memory at 4-bit | ~1.7 GB | ~1.3 GB |
| Context window | 1M tokens | 32K tokens |
| Input | text | text |
| Reasoning | Hybrid (thinking on/off) | Always thinks |
| License | Apache-2.0 | LFM Open License |
| API price / 1M tokens | Self-host only | Self-host only |
| Hugging Face downloads (30 days) | 2.7K | 8.1K |
Benchmarks
Higher is better. Bold marks the winner where both labs report a score. ≈ marks the index model's estimate for a score that wasn't reported, ✓ an independent measurement and * a score measured by another lab.
| Benchmark | Spark-X2.6 1.7B | LFM2.5 1.2B Thinking |
|---|---|---|
| MMLU-Pro Knowledge | ≈ 72 | 49.7 |
| GPQA Diamond Knowledge | 56.1 | 37.9 |
| Humanity's Last Exam Reasoning | 8.7 | ≈ 2 |
| AIME Math | 70.1 | 31.7 |
| SWE-bench Verified Coding | 59.8 | ≈ 12 |
| τ²-Bench Agentic | 85.3 | — |
| IFBench Instruction following | 65.5 | 44.9 |
| IFEval Instruction following | 89.4 | 88.4 |
| BFCL Agentic | 55.2 | 57.0 |
| MATH-500 Math | — | 88.0 |
| HMMT Math | 48.8 | — |
| AA-LCR Long context | 48.0 | — |
| SciCode Coding | 20.1 | — |
Scores are mostly reported by each lab with its own settings, so small gaps are noise. See the methodology. Open in the interactive comparison to add more models.
Questions
Which is better, Spark-X2.6 1.7B or LFM2.5 1.2B Thinking?
Spark-X2.6 1.7B scores higher on the Small LM Index: 50.1 vs 30.1. On the 5 benchmarks both labs report, Spark-X2.6 1.7B wins 4 and LFM2.5 1.2B Thinking wins 1.
Which is cheaper to run?
Neither model has a hosted API on OpenRouter, so you would run both yourself. Both need about 1.7 GB of memory at 4-bit quantization.
Which has the longer context window?
Spark-X2.6 1.7B has the longer context window: 1M tokens against 32K.
Can I use Spark-X2.6 1.7B and LFM2.5 1.2B Thinking commercially?
Spark-X2.6 1.7B is released under Apache-2.0 and LFM2.5 1.2B Thinking under LFM Open License. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.