LLM Hallucination Index RAG SpecialModel Insights
meta-llama-3-8b-instruct
Details
Developer
Meta
License
Llama 3
Model parameters
8b
Supported context length
8k
Price for prompt token
$0.2/Million tokens
Price for response token
$0.2/Million tokens
Model Performance Across Task-Types
Chainpoll Score
Short Context
0.89
Model Insights Across Task-Types
Digging deeper, here’s a look how meta-llama-3-8b-instruct performed across specific datasets
Short Context RAG
Performance Summary
| Tasks | Task insight | Cost insight | Dataset | Context adherence | Avg response length |
|---|---|---|---|---|---|
| Short context RAG | The model demonstrates good reasoning and comprehension skills, excelling at short context RAG. It also decent mathematical proficiency, as evidenced by its performance on DROP and ConvFinQA benchmarks. It comes out as one of the best small open source model ahead of mistral, qwen2 and Gemma. | One of the most affordable open source model with acceptable performance. It is nearly 15x cheaper than Sonnet 3.5 and 85% cheaper than Llama-3-70b. We highly recommend it if cost is your priority or task is simple. | Drop | 0.87 | 319 |
| Hotpot | 0.88 | 319 | |||
| MS Marco | 0.90 | 319 | |||
| ConvFinQA | 0.92 | 319 |