LLM Hallucination Index RAG SpecialModel Insights
qwen2-1.5b-instruct
Details
Developer
Alibaba
License
Tongyi Qianwen
Model parameters
1.5B
Supported context length
32k
Price for prompt token
$0.1/Million tokens
Price for response token
$0.1/Million tokens
Model Performance Across Task-Types
Chainpoll Score
Short Context
0.46
Model Insights Across Task-Types
Digging deeper, here’s a look how qwen2-1.5b-instruct performed across specific datasets
Short Context RAG
Performance Summary
| Tasks | Dataset | Context adherence | Avg response length |
|---|---|---|---|
| Short context RAG | Drop | 0.34 | 105.4 |
| Hotpot | 0.48 | 123 | |
| MS Marco | 0.70 | 243.2 | |
| ConvFinQA | 0.31 | 44.1 | |
| Medium context RAG | Medium context RAG | NaN | NA |
| Long context RAG | Long context RAG | NaN | NA |