A model update just shipped.
Do you know what it broke?

Model updates don't just improve — they redistribute capability. Some tasks get better; adjacent ones silently regress. Model-Drift Watch runs a fixed battery of production-shaped regression tasks against every major release, across providers, and publishes the diff — so you find out the same day, not from your customers.

Baseline v0.1 — cross-model scorecard (2026-07-13)

Workflow archetypeclaude-sonnet-5gemini-3.5-flashgpt-5.6-sol
Structured output (JSON)0/2525/2525/25
Long-doc extraction18/2525/2524/25
Instruction adherence25/2520/2524/25
Multilingual14/2525/2525/25
Tool/function calling25/2525/2525/25
Agentic multi-step25/2525/2525/25
Total107/150 · 71%145/150 · 97%148/150 · 99%

Headline: every claude-sonnet-5 failure here is one behavior — it wraps JSON in a markdown code fence despite an explicit "no markdown" instruction (43/43 fails; the JSON inside is otherwise correct). gemini-3.5-flash and gpt-5.6-sol honor the instruction 100%. A deterministic checker catches it every time.

Methodology: N=5 trials per task, 30 tasks, pinned seed, temperature 0 (or provider default), deterministic checkers; 0 infrastructure errors across all 450 trials. This is the inaugural baseline each future model release is diffed against. Task categories are published; raw task inputs never are.

  1. Every major release (~8–12/year across providers) we run the full battery within hours.
  2. Free scorecard published for every release — what changed, by archetype.
  3. Paid tier maps the diff to your workflow archetypes with patch recommendations, same day.

Get the next release diff, same day

Free forever for the public scorecard. No spam; one email per model release.