lfm-2.2-6B: benchmark results
Provider: Liquid AI. Access: Open.
Unified ELO 1281 ± 64, rank #1871 of 2133 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| 6G-Bench | 65.9 | Accuracy (%) on the full set of 6G-Bench's 3,722 expert-vali | 33.3 |
| 6G-Bench - Intent and Policy Reasoning | 76.4 | Accuracy (%) on the intent and policy reasoning tasks (T1, T | 29.6 |
| Persian Fluency Bench - Safety | 68.5 | Judge Score (%, judge google/gemini-2.5-flash) | 26.7 |
| 6G-Bench - AI-Native Networking and Agentic Control | 71.1 | Accuracy (%) on the AI-native networking and agentic control | 25.9 |
| 6G-Bench - Distributed Intelligence and Emerging Use Cases | 59.8 | Accuracy (%) on the distributed intelligence and emerging 6G | 25.9 |
| 6G-Bench - Network Slicing and Resource Management | 62.7 | Accuracy (%) on the network slicing and resource management | 25.9 |
| 6G-Bench - Trust, Security and SLA Awareness | 61 | Accuracy (%) on the trust, security and SLA awareness tasks | 23.1 |
| Persian Fluency Bench | 4.1 | Judge Score (%, judge google/gemini-2.5-flash) | 6.7 |
| Persian Fluency Bench - Conciseness | 4.1 | Judge Score (%, judge google/gemini-2.5-flash) | 6.7 |
| Persian Fluency Bench - Context Retention | 4.1 | Judge Score (%, judge google/gemini-2.5-flash) | 6.7 |
| Persian Fluency Bench - Grammar | 4.1 | Judge Score (%, judge google/gemini-2.5-flash) | 6.7 |
| Persian Fluency Bench - Idiomatic | 4.1 | Judge Score (%, judge google/gemini-2.5-flash) | 6.7 |
Interactive version: theaggregate.ai/model?slug=lfm-2-2-6b · How It Works · Data refreshed daily, snapshot 2026-10-11.