# Inception v3 CNN beats GPT-6 Astra and Claude Fable 5.1 on skin cancer test

Digest AI · Research · published 2026-10-25T17:15:38Z

Canonical: https://digestai.news/story/inception-v3-cnn-beats-gpt-6-astra-and-claude-fable-5-1-on-skin-cancer

## Summary

A researcher tested two frontier large language models, GPT-6 Astra and Claude Fable 5.1, against a fine-tuned version of Inception v3, a convolutional neural network published by Google in 2015. The goal was to diagnose 200 biopsy-proven skin lesions from the public MILK10k dataset. The CNN, fine-tuned on 68,011 images from the ISIC Archive on a single H100 GPU, correctly identified the diagnosis for 125 of the 200 lesions (62%). In comparison, GPT-6 Astra and Claude Fable 5.1 achieved accuracy rates of 42%, correctly naming the diagnosis for 84 and 85 lesions respectively.

The study highlights that specialized, smaller models can outperform general-purpose frontier LLMs in specific medical imaging tasks. The CNN was deliberately not state-of-the-art, using an outdated architecture, yet it outperformed the newer models on most categories, including melanoma and benign moles. The LLMs showed distinct biases, with GPT-6 Astra over-diagnosing basal cell carcinoma and Claude Fable 5.1 over-diagnosing actinic keratosis. When asked as a patient whether to see a doctor, both LLMs advised a visit in almost every case, regardless of the lesion's benign nature.

The author notes several caveats, including the small sample size of 200 lesions and the possibility that the LLMs may have seen the public dataset during training. The CNN is small enough to run in a phone browser using WebGPU, and all code, prompts, and results are available for public review. The analysis was assisted by Claude Code, one of the models tested.

## Key points

- Inception v3 CNN achieved 62% accuracy on 200 skin lesions, outperforming GPT-6 Astra and Claude Fable 5.1 at 42%.
- The CNN was fine-tuned on 68,011 images from the ISIC Archive and runs entirely in a phone browser using WebGPU.
- Both frontier LLMs advised seeing a doctor for 98-99 of 100 benign lesions when asked as a patient.

## Why it matters

This suggests that for specialized medical imaging tasks, small, fine-tuned, and potentially on-device models may be more accurate and efficient than general-purpose frontier LLMs, challenging the assumption that larger models are always superior for specific diagnostic applications.

## Sources

1. [A 10-year-old CNN outperforms frontier LLMs for skin cancer diagnosis](https://lynchlab.net/llm-skin-cancer) (lynchlab.net, 2026-10-25)

## Cite

Digest AI, "Inception v3 CNN beats GPT-6 Astra and Claude Fable 5.1 on skin cancer test", 25 October 2026, https://digestai.news/story/inception-v3-cnn-beats-gpt-6-astra-and-claude-fable-5-1-on-skin-cancer

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/inception-v3-cnn-beats-gpt-6-astra-and-claude-fable-5-1-on-skin-cancer.json