edit-drift
How much does an image model change what you told it to leave alone? Chained-edit benchmark for image models + installable scorer and Claude Code plugin.
Get picks like this daily. The day's top launches, AI/tech news, and a weekly opportunity spotlight — straight to your inbox.
This is 1 of 106 launches in ai benchmarking platforms — see how it stacks up on momentum and crowding →
90 other launches read as similar to this one →
Details
- External ID
- 1409109940
- Source
- GITHUB
- Company
- —
- Product
- edit-drift
- Website domain
- maxfusion.ai
- Launched
- Oct. 7, 2026
- Cohort
- —
- Upvotes
- 11
- Upvotes percentile
- 0.14
- Tags
- benchmark, claude-code, evaluation, flux, generative-ai, image-editing, image-generation, nano-banana
Enrichment
- Niche
- ai benchmarking platforms
- Vertical
- Horizontal
- Function
- Observability & eval
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Hobby / open-source project
- Normalized one-liner
- chained-edit drift benchmark for generative image models
- Manually corrected
- False
Could you build this?
Partial While the Claude Code plugin and wrapper CLI are simple to build, designing a rigorous, statistically sound benchmark dataset and perceptual drift scoring metric for generative diffusion models requires deep computer vision expertise.
What it would actually take: The system requires an orchestration harness in Python to chain multiple edit passes across various image diffusion APIs (Stable Diffusion, Midjourney, Flux) using structured mask/prompt pairs. The hard part is the drift evaluation metric: standard pixel metrics (MSE, PSNR) fail on generative outputs, requiring localized perceptual distance metrics (LPIPS, CLIP directional similarity, masked DINO embeddings) and an automated semantic segmentation pipeline to distinguish edited areas from unedited regions. Building a reliable, non-trivial benchmark requires expertise in computer vision evaluation methodologies.
Competitors
Other products that read as similar to this one — 90 launches clear the similarity bar, closest 8 shown.
Attention rank: #84 of 91 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 340 days after the earliest competitor.
- Drift · hn · 2026-06-10 · 6 upvotes · similarity 0.47
- claude-motion-skill · github · 2026-10-10 · 51 upvotes · similarity 0.41
- Parsimony · hn · 2026-10-13 · 41 upvotes · similarity 0.39
- lemo-wake · github · 2026-10-03 · 34 upvotes · similarity 0.38
- Mdarena · hn · 2026-04-05 · 22 upvotes · similarity 0.38
- claudebigimage · github · 2026-10-04 · 20 upvotes · similarity 0.37
- Paste-preview · hn · 2026-10-07 · 5 upvotes · similarity 0.36
- wmdrift · github · 2026-09-13 · 16 upvotes · similarity 0.36
Other launches for this product
- No other launches for this product.