AnthropicReleased Nov 2026Proprietary
Claude Fable 5.2 scores 86.5% overall on Roboflow's Vision Evals (November 10, 2026 snapshot), #2 of 67 models (Gemini 4 Argon leads at 87.5%).
Its best task is counting (86.5%, #1 of 67) and it ranks lowest on identification (96.9%, #7 of 67), a task most models score well on. It is a proprietary model from Anthropic, so commercial use follows Anthropic's terms.
Claude Fable 5.2Field median
Ranks out of 67 models (51 for OCR)
Not available to run yet
Claude Fable 5.2 is on the Vision Evals leaderboard. Running it in the Playground is not available yet. View evals
Vision Evals is Roboflow's ground-truth benchmark: every model runs the same real-world samples across five vision tasks, and answers are scored against ground truth.
Evals updated November 10, 2026Pricing updated November 10, 2026
Strengths
Weak spot
Above the field median, but its lowest rank.
Claude Fable 5.2Field median
Ranks out of 67 models (51 for OCR)
| Task | Score | Field (0 to 100) | Rank | Cost / sample | Speed |
|---|---|---|---|---|---|
| Object Detection (low) | 81.6% ±0.3, Mean of 3 runs, range 81.2 to 81.9 | #4 of 67 | $0.048 | 13.25s | |
| Object Detection (high) | 83.3% ±0.5, Mean of 3 runs, range 82.8 to 83.7 | #4 of 35 | $0.055 | 14.79s | |
| Counting (low) | 86.5% ±0.0, Mean of 3 runs, range 86.5 to 86.5 | #1 of 67 | $0.015 | 6.42s | |
| Counting (high) | 92.8% ±1.4, Mean of 3 runs, range 91.9 to 94.6 | #1 of 35 | $0.021 | 8.08s | |
| Identification (low) | 96.9% ±0.0, Mean of 3 runs, range 96.9 to 96.9 | #7 of 67 | $0.011 | 5.60s | |
| Identification (high) | 94.8% ±1.6, Mean of 3 runs, range 93.8 to 96.9 | #9 of 35 | $0.012 | 5.98s | |
| OCR (low) | 79.5% | #3 of 51 | $0.030 | 9.03s | |
| |||||
| OCR (high) | 84.8% | #2 of 51 | $0.043 | 12.14s | |
| |||||
| Reasoning (low) | 88.1% ±1.3, Mean of 3 runs, range 86.8 to 89.4 | #2 of 67 | $0.016 | 6.28s | |
| Reasoning (high) | 92.0% ±1.0, Mean of 3 runs, range 91.4 to 93.4 | #1 of 54 | $0.024 | 8.48s | |
Overall benchmark score against estimated cost per sample, on a log scale. Upper-left is the sweet spot: high quality at low cost.
66 of 67 models plotted (the rest have no measured figure for this axis) · scores and efficiency pooled across all five tasks at low effort · Claude Fable 5.2 highlighted
Claude Fable 5.2 scores are the mean of 3 runs per task at both low and high effort; OCR is a single preliminary run · Methodology
View all Vision Evals →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.
Claude Fable 5.2 costs $8.00 per 1M input tokens and $40.00 per 1M output tokens.
Pricing updated Nov 10, 2026
Claude Fable 5.2 is proprietary: the weights are not distributed, and the Claude Fable 5.2 license is the vendor's commercial terms of service that you accept when you call the API.
Vendor terms govern data retention, whether your inputs can be trained on, rate limits, and regional availability, and they can change with notice. Review them if you handle regulated or customer data.
Proprietary terms are set by the vendor rather than negotiated per project, and no open-source obligation attaches to your code. If you would rather deploy an open model instead of Claude Fable 5.2, Roboflow's licensing page lists the models whose commercial license is included when you deploy on Roboflow.
Do not hesitate to reach out with questions for your commercial project. Our team will help you start solving business problems on the first call. See Roboflow commercial licensing for the models included in each plan.
Talk to salesLicense information is provided as a guide and is not legal advice.
Other versions in the same family as Claude Fable 5.2.
Yes. Claude Fable 5.2 accepts image input and handles OCR, object counting, identification, visual reasoning, and object detection. On Roboflow's Vision Evals its strongest task is Counting at 86.5% (#1 of 67 at low effort).
Yes, and it is one of the model's strongest vision skills: it scores 79.5% (#3 of 51 at low effort) on Vision Evals OCR, which covers single readings, full transcriptions, structured JSON, and locating text with boxes. Within that, full transcriptions score 94.7%, and returning the fields of a label or form as JSON scores 92.8%.
Yes. On Vision Evals, Claude Fable 5.2 scores 81.6% mAP@50 on object detection (#4 of 67 at low effort) and 86.5% judge-graded accuracy on object counting.
On our benchmark's task mix, Claude Fable 5.2 averages $0.03 per sample at $8.00 per 1M input and $40.00 per 1M output tokens (#63 of 67 on cost), with an average speed of 9.3s per sample across the benchmark. Actual cost depends on your images and prompts.
On the overall Vision Evals ranking, Claude Fable 5.2 sits #2 of 67 at 86.5%, just behind Gemini 4 Argon (87.5%) and just ahead of GPT-6.1 Terra (85.8%). See the full side-by-side: Claude Fable 5.2 vs Gemini 4 Argon.