Mistral Large 4 vs SAM 3
Compare Mistral Large 4 and SAM 3 side-by-side. See how these vision models stack up in Object Detection.
Compare Mistral Large 4 vs SAM 3 live
Run the same image across every model that supports a task and compare their outputs side-by-side.
Detect and compare bounding boxes across models on the same image.
Upload an image
Drag and drop an image here, or click to browse
Models in this comparison
Mistral Large 4 vs SAM 3 Comparison Table
Evals updated October 8, 2026Pricing updated October 8, 2026
| Property | Mistral Large 4 | SAM 3 |
|---|---|---|
| Organization | Mistral | Meta |
| Category | open | open |
| Modality | multimodal | multimodal |
| Release Date | Oct 2026 | Nov 2025 |
| Context Window | 1.0M | — |
| Parameters | 1.05T total, 49B active | Unknown |
| License | Custom | Custom |
| Pricing per 1M tokens | ||
| Input $/1M | $0.680 | No published price |
| Output $/1M | $2.09 | No published price |
| Vision Tasks | ||
| Object Detection | Demo | Demo |
| Captioning | Demo | Not listed |
| Chart Question Answering | Supported | Not listed |
| Classification | Demo | Not listed |
| Document Question Answering | Supported | Not listed |
| Image Tagging | Supported | Not listed |
| Instance Segmentation | Not listed | Supported |
| Multi-Label Classification | Supported | Not listed |
| OCR | Demo | Not listed |
| Open Vocabulary Object Detection | Not listed | Supported |
| Phrase Grounding | Supported | Not listed |
| Promptable Concept Segmentation | Not listed | Demo |
| Video Object Tracking | Not listed | Supported |
| Vision Language | Supported | Not listed |
| Visual Question Answering | Demo | Not listed |
| Zero Shot Segmentation | Not listed | Supported |
| Model Features | ||
| Foundation Vision | Supported | Supported |
| Multimodal Vision | Supported | Supported |
| LLMs with Vision Capabilities | Supported | Not listed |
| Zero-shot Detection | Not listed | Supported |
Vision Evalsground-truth scores across 6 vision tasks, pooled at low effort | ||
| Overall | 68.5% | Not evaluated |
| Avg cost / sample | $0.0018 | – |
| Avg speed / sample | 8.78s | – |
| By task | ||
| Object Detection (low) | 59.3% ±0.7, Mean of 3 runs, range 58.5 to 60.0 | – |
| Object Detection (high) | 50.2% ±2.5, Mean of 3 runs, range 48.0 to 53.0 | – |
| Counting (low) | 54.5% ±0.7, Mean of 3 runs, range 54.0 to 55.4 | – |
| Counting (high) | 63.1% ±2.0, Mean of 3 runs, range 60.8 to 64.9 | – |
| Identification (low) | 85.4% ±3.1, Mean of 3 runs, range 81.3 to 87.5 | – |
| Identification (high) | 85.4% ±1.6, Mean of 3 runs, range 84.4 to 87.5 | – |
| OCR (low) | 92.7% ±0.8, Mean of 3 runs, range 91.8 to 93.3 | – |
| OCR (high) | 87.1% ±4.4, Mean of 3 runs, range 81.6 to 90.4 | – |
| Data Extraction (low) | 80.1% ±1.0, Mean of 3 runs, range 79.4 to 81.4 | – |
| Data Extraction (high) | 82.5% ±1.0, Mean of 3 runs, range 81.4 to 83.5 | – |
| Reasoning (low) | 38.9% ±0.3, Mean of 3 runs, range 38.4 to 39.1 | – |
| Reasoning (high) | 57.6% ±2.0, Mean of 3 runs, range 55.6 to 59.6 | – |
Mistral Large 4 vs SAM 3: Overview
Mistral Large 4, nicknamed Le Chonk, is a natively multimodal mixture-of-experts model from Mistral that accepts interleaved text and image input and produces text output. It uses a granular MoE design with roughly 1.05 trillion total parameters and 49 billion active per token, reported as 52 billion when embeddings and output layers are counted, paired with a 1.6 billion parameter vision encoder and a context window of one million tokens. The model is trained from scratch on about 3,800 NVIDIA Grace Blackwell GPUs in Mistral's European data centers and supports more than 160 languages. It behaves as a hybrid instruct and reasoning system, with a reasoning effort setting that selects between direct answers and longer deliberation, alongside function calling and structured output for agentic workflows.
Image understanding is a focus of this generation, covering documents, charts, technical drawings and natural scenes, and the model emits bounding box coordinates for visual grounding queries. Reported grounding results include 42 percent on Dense200 and 73 percent on the DIOR-RSVG remote sensing benchmark. Mistral describes agentic vision workflows in which the model zooms into gigapixel satellite imagery or engineering drawings to verify details, and reports coding results such as 62 percent on DeepSWE. Figures published at preview time are preliminary because the reinforcement learning phase is still in progress.
Released on November 19th, 2025, Segment Anything 3 (SAM 3) is a zero-shot image segmentation model that “detects, segments, and tracks objects in images and videos based on concept prompts.” This model was developed by Meta as the third model in the Segment Anything series.
Unlike its previous SAM models (Segment Anything and Segment Anything 2), you can provide SAM 3 with the prompt “shipping container” and it will generate precise segmentation masks for all shipping containers in an image. SAM 3 generates segmentation masks that correspond to the location of the objects found with a text prompt.
Frequently Asked Questions
SAM 3 has not yet been evaluated on Roboflow's current Vision Evals, so this comparison shows specs, licensing, and pricing rather than benchmark scores.
Yes. The comparison demo on this page runs both models on the same image side by side for object detection in the free Roboflow Playground. You can try it instantly, and a free account unlocks unlimited runs.
