Open-Ended Visual Question Answering on VizWiz (val)
56.39AccuracyLlama-2 Chat 7B
Evaluation results
| Method | Links | |
|---|---|---|
| Llama-2 Chat 7BLLM backbone=Llama-2, Model Size=7B, Variant=Chat2025.12 | 56.39 | |
| Llama-2 7BLLM backbone=Llama-2, Model Size=7B2025.12 | 55.98 | |
| Mistral v0.1 7BLLM backbone=Mistral, Model Size=7B, Variant=v0.12025.12 | 55.32 | |
| Mistral Instruct v0.1 7BLLM backbone=Mistral, Model Size=7B, Variant=Instruct v0.12025.12 | 54.35 | |
| LLaVa v1.5 7B (Base)LLM backbone=LLaVa v1.5, Model Size=7B, Variant=Base2025.12 | 54.25 | |
| Moxin-7BLLM backbone=Moxin-7B, Model Size=7B2025.12 | 54.08 |