Claude Fable 5.1 vs Gemma 4 31B
Compare Claude Fable 5.1 and Gemma 4 31B side-by-side. See how these vision models stack up in Object Detection, OCR, Image Captioning, Open Prompt, and Classification.
Compare Claude Fable 5.1 vs Gemma 4 31B 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
Claude Fable 5.1 vs Gemma 4 31B on Vision Evals
Claude Fable 5.1 scores higher on all 4 Vision Evals tasks.
The widest gap is Reasoning, where Claude Fable 5.1 leads 72.0% to 52.8%.
Overall, Claude Fable 5.1 averages 74.7% (#14 of 63) against 57.7% (#39 of 63) for Gemma 4 31B.
Gemma 4 31B is cheaper ($0.0015 vs $0.039 per sample), while Claude Fable 5.1 is faster (8.3s vs 34.4s per sample).
Claude Fable 5.1 vs Gemma 4 31B Comparison Table
Evals updated October 13, 2026Pricing updated October 17, 2026
| Property | Claude Fable 5.1 | Gemma 4 31B |
|---|---|---|
| Organization | Anthropic | |
| Category | closed | open |
| Modality | multimodal | multimodal |
| Release Date | Sep 2026 | Apr 2026 |
| Context Window | 1.0M | 256K |
| Parameters | Unknown | 31B |
| License | Proprietary | Apache 2.0 |
| Commercial use | commercial use per vendor terms | free for commercial use |
| Pricing per 1M tokens | ||
| Input $/1M | $10.00 | $0.200 |
| Output $/1M | $50.00 | $0.400 |
| Vision Tasks | ||
| Captioning | Demo | Demo |
| Chart Question Answering | Supported | Supported |
| Classification | Demo | Demo |
| Document Question Answering | Supported | Supported |
| Image Tagging | Supported | Supported |
| Multi-Label Classification | Supported | Supported |
| Object Detection | Demo | Demo |
| OCR | Demo | Demo |
| Vision Language | Supported | Supported |
| Visual Question Answering | Demo | Demo |
| Model Features | ||
| Foundation Vision | Supported | Supported |
| LLMs with Vision Capabilities | Supported | Supported |
| Multimodal Vision | Supported | Supported |
Vision Evalsground-truth scores across 5 vision tasks, pooled at low effort | ||
| Overall | 74.7% | 57.7% 4/5 tasks |
| Quantizationsself-hosted | ||
| Avg cost / sample | $0.039 | $0.0015 |
| Avg speed / sample | 8.35s | 34.36s |
| By task | ||
| Object Detection (low) | 61.4% ±0.5, Mean of 3 runs, range 61.0 to 62.0 | 47.5% ±0.6, Mean of 3 runs, range 46.8 to 48.0 |
| Object Detection (high) | 65.0% ±0.4, Mean of 3 runs, range 64.6 to 65.3 | – |
| Counting (low) | 69.4% ±2.7, Mean of 3 runs, range 66.2 to 71.6 | 51.4% ±2.7, Mean of 3 runs, range 48.6 to 54.0 |
| Counting (high) | 73.0% ±4.7, Mean of 3 runs, range 67.6 to 77.0 | – |
| Identification (low) | 97.9% ±1.6, Mean of 3 runs, range 96.9 to 100.0 | 79.2% ±1.6, Mean of 3 runs, range 78.1 to 81.3 |
| Identification (high) | 96.9% ±3.1, Mean of 3 runs, range 93.8 to 100.0 | – |
| OCR (low) | 72.9% | – |
| by category |
| |
| OCR (high) | 75.3% | – |
| by category |
| |
| Reasoning (low) | 72.0% ±1.3, Mean of 3 runs, range 70.9 to 73.5 | 52.8% ±1.3, Mean of 3 runs, range 51.7 to 54.3 |
| Reasoning (high) | 73.1% ±2.0, Mean of 3 runs, range 71.5 to 75.5 | – |
Claude Fable 5.1 vs Gemma 4 31B: Overview
Claude Fable 5.1 is a proprietary multimodal model from Anthropic in the Mythos-class tier of the Claude 5 family, positioned above Claude Opus for demanding reasoning and long-horizon agentic work. It accepts text and images as input and returns text, with a one million token context window and a maximum output of 128 thousand tokens. Adaptive thinking is always on, and an effort parameter controls how much reasoning the model applies to a given request. Anthropic reports a reliable knowledge and training data cutoff of June 2026. Claude Fable 5.1 and Claude Mythos 5.1 share the same underlying model; the difference between them is the set of safety classifiers applied to dual-use cybersecurity and biology requests.
On the vision side, Anthropic documents improvements in reading dense charts, financial filings, and tables nested inside PDF documents, which extends the model toward document understanding, chart question answering, and spreadsheet and slide work. Reported evaluations cover agentic scientific research on Terminal-Bench-Science 0.1, agentic coding on Terminal-Bench 4.0, computer use on OSWorld 2.0, and multidisciplinary reasoning on Humanity's Last Exam. Model weights are not published.
Gemma 4 31B is the largest dense model in Google's Gemma 4 family, built from the same research as Gemini 3 and released as open weights under the Apache 2.0 license. It supports a 256K token context window with text and image input, configurable thinking mode for step-by-step reasoning, and multilingual support across 140+ languages. The unquantized model fits on a single 80GB GPU.
For vision tasks, Gemma 4 31B supports image understanding with variable aspect ratios and resolutions, and can output structured bounding boxes for UI element detection, making it useful for document parsing and UI understanding. Compared to Gemma 3, it delivers stronger reasoning and multimodal performance. It is part of a four-size family alongside the 26B A4B MoE variant and two on-device models (E2B, E4B), with the 31B dense variant optimized for output quality and fine-tuning over inference speed.
Frequently Asked Questions
On Roboflow's Vision Evals, Claude Fable 5.1 performed better. It scores higher on all 4 vision tasks and averages 74.7% (#14 of 63) against 57.7% (#39 of 63) for Gemma 4 31B. These scores measure vision capabilities only; either model may perform differently on work outside this benchmark's tasks.
Yes. On the Vision Evals Reasoning benchmark at low effort, Claude Fable 5.1 leads with 72.0% against 52.8%. This is the widest gap between the two models across the benchmark's tasks.
Gemma 4 31B is cheaper to run. Across Roboflow's Vision Evals task mix it averaged $0.0015 per sample against $0.039. Actual costs depend on your image sizes, prompts, and output length.
Claude Fable 5.1 is faster. Across Roboflow's Vision Evals it averaged 8.3s per inference against 34.4s. Latency varies with image size, prompt length, and provider load, so treat these as relative rather than guaranteed figures.
Yes. The comparison demo on this page runs both models on the same image side by side for object detection and OCR in the free Roboflow Playground. You can try it instantly, and a free account unlocks unlimited runs.