Claude Fable 5.2 vs Gemma 4 31B
Claude Fable 5.2 averages 86.5% across five Vision Evals tasks and Gemma 4 31B 57.7% across four (#2 and #43 of 67 models), so Claude Fable 5.2 leads overall.
- Claude Fable 5.2 wins all 4 tasks both ran, with its widest lead on reasoning (88.1% vs 52.8%).
- Gemma 4 31B is under a tenth of the cost: $0.0015 vs $0.031 overall average cost per sample.
- Claude Fable 5.2 is faster: 9.3s vs 34.4s per sample.
Roboflow Vision Evals. Full breakdown below.
Compare Claude Fable 5.2 vs Gemma 4 31B 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.
Claude Fable 5.2 vs Gemma 4 31B on Vision Evals
Roboflow Vision Evals. Full breakdown below.
Claude Fable 5.2 vs Gemma 4 31B Comparison Table
Evals updated November 10, 2026Pricing updated November 10, 2026
| Property | Claude Fable 5.2 | Gemma 4 31B |
|---|---|---|
| Organization | Anthropic | |
| Category | closed | open |
| Modality | multimodal | multimodal |
| Release Date | Nov 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 | $8.00 | $0.200 |
| Output $/1M | $40.00 | $0.400 |
| Vision Tasks | ||
| Captioning | Supported | Demo |
| Classification | Supported | Demo |
| Object Detection | Supported | Demo |
| OCR | Supported | Demo |
| Visual Question Answering | Supported | Demo |
| Model Features | ||
Both support 3 more features | ||
Vision Evalsground-truth scores across 5 vision tasks, pooled at low effort | ||
| Overall | 86.5% | 57.7%* |
| Quantizationsself-hosted | ||
| Avg cost / sample | $0.031 | $0.0015 |
| Avg speed / sample | 9.33s | 34.36s |
| By task | ||
| Object Detection (low) | 81.6% ±0.3, Mean of 3 runs, range 81.2 to 81.9 | 47.5% ±0.6, Mean of 3 runs, range 46.8 to 48.0 |
| Object Detection (high) | 83.3% ±0.5, Mean of 3 runs, range 82.8 to 83.7 | – |
| Counting (low) | 86.5% ±0.0, Mean of 3 runs, range 86.5 to 86.5 | 51.4% ±2.7, Mean of 3 runs, range 48.6 to 54.0 |
| Counting (high) | 92.8% ±1.4, Mean of 3 runs, range 91.9 to 94.6 | – |
| Identification (low) | 96.9% ±0.0, Mean of 3 runs, range 96.9 to 96.9 | 79.2% ±1.6, Mean of 3 runs, range 78.1 to 81.3 |
| Identification (high) | 94.8% ±1.6, Mean of 3 runs, range 93.8 to 96.9 | – |
| OCR (low) | 79.5% | – |
| by category |
| |
| OCR (high) | 84.8% | – |
| by category |
| |
| Reasoning (low) | 88.1% ±1.3, Mean of 3 runs, range 86.8 to 89.4 | 52.8% ±1.3, Mean of 3 runs, range 51.7 to 54.3 |
| Reasoning (high) | 92.0% ±1.0, Mean of 3 runs, range 91.4 to 93.4 | – |
Claude Fable 5.2 vs Gemma 4 31B: Overview
Claude Fable 5.2 is a proprietary multimodal model from Anthropic, released on November 9, 2026. It accepts text and images and produces text, with a one million token context window and up to 128,000 output tokens. Adaptive thinking is always enabled, with an effort setting that controls reasoning. Anthropic positions Fable for demanding reasoning and long-running technical work, including coding and analysis of visual inputs.
Fable 5.2 succeeds Fable 5.1. Anthropic reports improvements in coding, reasoning, and vision, with output generation up to 30% faster than its predecessor. Fable and Mythos 5.2 share an underlying model, with additional cybersecurity and biology safeguards applied to Fable. Model weights are not publicly released.
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 5.2 performed better. It scores higher on all 4 vision tasks and averages 86.5% (#2 of 67) against 57.7% (#43 of 67) 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 5.2 leads with 88.1% 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 $0.031. Actual costs depend on your image sizes, prompts, and output length.
Claude Fable 5.2 is faster. Across Roboflow's Vision Evals it averaged 9.3s per inference against 34.4s. Latency varies with image size, prompt length, and provider load, so treat these as relative rather than guaranteed figures.