LLaVA-1.5 vs Qwen3.5 9b
Qwen3.5 9b: 56.0% average across four Vision Evals tasks (#44 of 65 models). We have not run LLaVA-1.5 through Roboflow's Vision Evals, so there is no head-to-head score.
Compare LLaVA-1.5 vs Qwen3.5 9b live
Run the same image across every model that supports a task and compare their outputs side-by-side.
These models don't share enough common tasks for a side-by-side demo. See the comparison table below for their capabilities.
LLaVA-1.5 vs Qwen3.5 9b Comparison Table
Evals updated October 26, 2026Pricing updated November 5, 2026
| Property | LLaVA-1.5 | Qwen3.5 9b |
|---|---|---|
| Organization | Microsoft | Qwen |
| Category | open | open |
| Modality | multimodal | multimodal |
| Release Date | Oct 2023 | Mar 2026 |
| Context Window | — | 262K |
| Parameters | 7B, 13B | 9B |
| License | Custom | Apache 2.0 |
| Commercial use | See license terms | free for commercial use |
| Pricing per 1M tokens | ||
| Input $/1M | No published price | $0.170 |
| Output $/1M | No published price | $0.250 |
| Vision Tasks | ||
| Visual Question Answering | Supported | Demo |
| Captioning | Not listed | Demo |
| Chart Question Answering | Not listed | Supported |
| Classification | Not listed | Supported |
| Document Question Answering | Not listed | Supported |
| Image Tagging | Not listed | Supported |
| Multi-Label Classification | Not listed | Supported |
| Object Detection | Not listed | Supported |
| OCR | Not listed | Demo |
Both support 1 more task | ||
| Model Features | ||
| Foundation Vision | Not listed | Supported |
Both support 2 more features | ||
Vision Evalsground-truth scores across 5 vision tasks | ||
| Overall | Not evaluated | 56.0%* |
| Quantizationsself-hosted | ||
| Avg cost / sample | – | $0.0021 |
| Avg speed / sample | – | 41.36s |
| By task | ||
| Object Detection | – | 38.1% ±5.7, Mean of 3 runs, range 33.5 to 44.9 |
| Counting | – | 56.8% ±1.4, Mean of 3 runs, range 55.4 to 58.1 |
| Identification | – | 83.3% ±1.6, Mean of 3 runs, range 81.3 to 84.4 |
| OCR | – | – |
| by category | ||
| Reasoning | – | 45.9% ±1.7, Mean of 3 runs, range 44.4 to 47.7 |
LLaVA-1.5 vs Qwen3.5 9b: Overview
LLaVA-1.5 is an open-source large multimodal model released in October 2023 by researchers at the University of Wisconsin-Madison and Microsoft Research. It builds on the original LLaVA architecture by introducing targeted refinements: switching the vision encoder to CLIP-ViT-L at 336-pixel resolution, replacing the projection layer with a two-layer MLP, and adding academic-task-oriented visual question answering data with response formatting prompts during training. These modifications achieve state-of-the-art performance across 11 benchmarks at release, with training completing in approximately one day on a single 8-A100 node.
The model accepts an image paired with a text prompt and generates natural language responses, supporting visual question answering, image captioning, and open-ended visual conversation. LLaVA-1.5 is available in 7B and 13B parameter variants built on the Vicuna language model, and is distributed under the Llama 2 Community License due to its Llama-2-based foundation. The original LLaVA paper was presented as an oral at NeurIPS 2023. Subsequent releases in the series (LLaVA-NeXT (LLaVA-1.6), LLaVA-NeXT-Video, and LLaVA-OneVision) are separate models with their own release pages and build on this foundation with expanded OCR, video, and multi-image capabilities.
Qwen3.5-9B is a 9-billion-parameter multimodal foundation model developed by Alibaba Cloud's Qwen team, released on March 2, 2026 as part of the Qwen3.5 model family.
The model supports vision-language inputs through an early-fusion multimodal architecture built on a dense hybrid foundation of Gated Delta Networks and Gated Attention. It can also operate in a text-only mode by skipping the vision encoder during inference. It provides a 262,144-token context window (extensible to ~1M tokens via YaRN) and is released under the Apache License 2.0. Within the current AI landscape, Qwen3.5-9B offers a strong balance of capability and efficiency, making it well-suited for multimodal assistants, document analysis, long-context reasoning, and developer-deployed agentic systems.