gemini-2.5-flash dropped on common-sense-1. gpt-5.4-mini failing spatial-1. gemini-2.5-pro recovering. gpt-5.4-mini, gpt-5.4-mini scores rising.
April 15, 2026 — 12:30 PM CT
Drift Alerts
- SCORE_RISE openai/gpt-5.4-mini on math-1
- SCORE_RISE openai/gpt-5.4-mini on spatial-1
- IMPROVEMENT gemini/gemini-2.5-pro on common-sense-1
- SCORE_DROP gemini/gemini-2.5-flash on common-sense-1
Scorecard
| Model | ambiguity-1 | causality-1 | code-1 | common-sense-1 | logic-1 | math-1 | spatial-1 |
|---|---|---|---|---|---|---|---|
| anthropic/claude-haiku-4-5 | ✓ (4.5) | ✓ (4.67) | ✓ (4.67) | ✓ (3.33) | ✓ (5) | ✓ (5) | ✓ (5) |
| anthropic/claude-opus-4-6 | ✓ (5) | ✓ (5) | ✓ (4.8) | ✓ (4.33) | ✓ (5) | ✓ (5) | ✓ (5) |
| anthropic/claude-sonnet-4-6 | ✓ (4.5) | ✓ (4.83) | ✓ (4.4) | ✓ (4.33) | ✓ (4.83) | ✓ (5) | ✓ (5) |
| gemini/gemini-2.5-flash | ✓ (4.5) | ✓ (3.67) | ✓ (4.5) | ✓ (2.8)was 4.5 | ✓ (5) | ✓ (5) | ✓ (5) |
| gemini/gemini-2.5-pro | ✓ (4.33) | ✓ (4.83) | ✓ (4.83) | ✓ (5)was ✗ () | ✓ (5) | ✓ (5) | ✓ (5) |
| ollama/llama3 | — | — | — | — | — | — | — |
| openai/gpt-5.4 | ✓ (4.5) | ✓ (4.83) | ✓ (4.83) | ✓ (4.33) | ✓ (4.83) | ✓ (4.6) | ✓ (5) |
| openai/gpt-5.4-mini | ✓ (5) | ✓ (4.83) | ✓ (4.8) | ✓ (4) | ✓ (5) | ✓ (5)was 4 | ✗ (3.67)was 2.4 |
Model Status
- → anthropic/claude-haiku-4-5 stable
- → anthropic/claude-opus-4-6 stable
- → anthropic/claude-sonnet-4-6 stable
- ↓ gemini/gemini-2.5-flash down
- ↑ gemini/gemini-2.5-pro up
- → openai/gpt-5.4 stable
- ↑ openai/gpt-5.4-mini up
Raw Data
- Detail log — full responses and judge verdicts per prompt
- JSON — structured data for programmatic access
- Markdown — plain text report
- responses.json — raw model outputs
- judgments.json — raw judge verdicts
- run.log — debug log
- Agent Skill — how to read and interpret this data
- Methodology — how evaluations work