Mistral Small 3.1 24B vs Qwen3.5 397B A17B
Compare Mistral Small 3.1 24B and Qwen3.5 397B A17B side-by-side. See how these vision models stack up in Image Captioning, OCR, and Open Prompt.
Compare Mistral Small 3.1 24B vs Qwen3.5 397B A17B live
Run the same image across every model that supports a task and compare their outputs side-by-side.
Extract and compare text from images across multiple models.
Upload an image
Drag and drop an image here, or click to browse
Models in this comparison
Mistral Small 3.1 24B vs Qwen3.5 397B A17B: Overview
Mistral Small 3.1 24B, released on March 17, 2025, is an open-weight multimodal model from Mistral AI, distributed under the Apache-2.0 license. With around 24B parameters and a 128K token context window, it is available in both base and instruction-tuned (“Instruct”) variants. The model introduces vision support alongside text, enabling tasks like multimodal reasoning, captioning, and image-based Q&A.
It is multilingual, supporting many languages, and is optimized for fast responses, function calling, structured dialogue, and long-context reasoning. Despite its size, the model can be run locally in quantized formats, fitting on machines with ~32GB RAM, making it accessible to developers outside large cloud setups. However, the output length is smaller than the 128K input window, meaning long generations may require chaining. In addition, using full vision features or the maximum context window significantly increases compute costs, and performance on highly complex reasoning or enterprise-scale tasks still trails larger proprietary frontier models.
Qwen3.5-397B-A17B is a 397B-parameter (17B active) open-weight multimodal model developed by Alibaba’s Qwen team, released on 2026-02-16 under Apache-2.0. It supports text and image inputs with text outputs, combining a sparse Mixture-of-Experts architecture with Gated Delta Networks for efficient scaling. The model provides native vision-language reasoning and a large ~262K token context window, extendable to ~1M tokens.
As the first open-weight release in the Qwen3.5 family, it positions itself as a high-capacity, long-context alternative in the large vision-language space, balancing scale and efficiency via sparse activation. It is designed for advanced reasoning, coding, agent workflows, and multimodal understanding tasks.
Mistral Small 3.1 24B vs Qwen3.5 397B A17B Comparison Table
| Property | Mistral Small 3.1 24B | Qwen3.5 397B A17B |
|---|---|---|
| Organization | Mistral | Qwen |
| Category | open | open |
| Modality | multimodal | multimodal |
| Release Date | Mar 2025 | Feb 2026 |
| Context Window | 128K | 262K |
| Parameters | 24B | 397B |
| License | Apache 2.0 | Apache 2.0 |
| Pricing per 1M tokens | ||
| Input $/1M | $0.351 | $0.385 |
| Output $/1M | $0.555 | $2.45 |
| Vision Tasks | ||
| Captioning | Demo | Demo |
| OCR | Demo | Demo |
| Vision Language | ||
| Visual Question Answering | Demo | Demo |
| Object Detection | ||
| Model Features | ||
| Multimodal Vision | ||
| LLMs with Vision Capabilities | ||
Vision Evalspass/fail results · 67 prompts Score key:≥75%40–74%<40% | ||
| Visual Understanding | ||
| Overall Score | 58.21% | |
| Avg Response Time | 56.61s | |
| Median input tokensincl. image tokens | 1.1K | |
| Median output tokens | 54 | |
| Est. cost / taskon this benchmark | $0.0006 | |
| Defect Detection | 66.7%(10/15) | |
| Document Understanding | 77.8%(7/9) | |
| Object Counting | 20%(2/10) | |
| Object Understanding | 64.3%(9/14) | |
| Spatial Understanding | 57.9%(11/19) | |
| OCR | ||
| Overall Score | 68.56% | |
| Avg Response Time | 7.45s | |
| Median input tokensincl. image tokens | 122 | |
| Median output tokens | 20 | |
| Est. cost / taskon this benchmark | $0.0001 | |
| Focused Scene OCR | 57.6%(57/99) | |
| Handwritten Math | 80%(8/10) | |
| License Plate Recognition | 100%(30/30) | |
| Text Recognition | 70%(21/30) | |
| VQA & Extraction | 68.3%(41/60) | |
Output tokens (incl. reasoning) and est. cost / task are measured on this benchmark from a single low-temperature run, and shown only for models whose run covered at least 90% of prompts. Methodology