LiT Qwen3-30B-A3B Instruct (checkpoint unspecified): benchmark results
Provider: Other.
Aggregate Rating 1494 ± 41, rank #2459 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LiT (Lost in Translation) - STEM Abstracts | 19.4 | MQM>=80 rate (%): share of round-trip translations whose bac | 9.5 |
| LiT (Lost in Translation) | 21.4 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) - Core Semantics | 30.7 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) - Discourse Coherence | 28 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) - Humanities Abstracts | 14.4 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) - Implicit Content | 23.2 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) - Informal Text | 3.4 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) - Pragmatic Inference | 25.5 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) - Social Interaction | 26.8 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) Robustness | 14 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) Robustness - Conceptual and Abstract Nuance | 25 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
| LiT (Lost in Translation) Robustness - Idioms and Cultural Metaphors | 7.3 | MQM>=80 rate (%): share of round-trip translations whose bac | 4.8 |
Interactive version: theaggregate.ai/model?slug=lit-qwen3-30b-a3b-instruct-checkpoint-unspecified · How It Works · Data refreshed daily, snapshot 2026-11-08.