gpt-5.4-mini lost spatial-1. claude-haiku-4-5, gemini-2.5-flash recovering.
April 25, 2026 — 5:15 PM CT
Drift Alerts
- REGRESSION openai/gpt-5.4-mini on spatial-1
- IMPROVEMENT anthropic/claude-haiku-4-5 on common-sense-1
- IMPROVEMENT gemini/gemini-2.5-flash on causality-1
Provider Status
- OpenAI Elevated errors on model gpt-5.5 in codex
- Anthropic Investigated elevated errors and slower responses on claude.ai
- Anthropic Elevated errors on Claude Opus 4.7
- Anthropic Elevated errors on Claude Opus 4.7
- Anthropic Claude Code v2.1.120 Crashes on Startup
- Anthropic Elevated error rates on Claude Opus 4.7
- Anthropic Issues with sign-ups on platform.claude.com
- Anthropic Elevated errors on Claude Opus 4.7
- Anthropic Elevated errors on Sonnet 4.6
Scorecard
| Model | ambiguity-1 | causality-1 | code-1 | common-sense-1 | logic-1 | math-1 | spatial-1 |
|---|---|---|---|---|---|---|---|
| anthropic/claude-haiku-4-5 | ✓ (4.33) | ✓ (4.67) | ✓ (4.5) | ✓ (3.33)was ✗ (3.17) | ✓ (5) | ✓ (5) | ✓ (5) |
| anthropic/claude-opus-4-6 | ✓ (5) | ✓ (4.67) | ✓ (4.83) | ✓ (4.33) | ✓ (5) | ✓ (5) | ✓ (5) |
| anthropic/claude-sonnet-4-6 | ✓ (4.67) | ✓ (4.5) | ✓ (4.5) | ✓ (3.5) | ✓ (4.67) | ✓ (5) | ✓ (5) |
| gemini/gemini-2.5-flash | ✓ (4.5) | ✓ (3.5)was ✗ (3.5) | ✓ (5) | ✓ (4.67) | ✓ (5) | ✓ (5) | ✓ (5) |
| gemini/gemini-2.5-pro | ✓ (4.5) | ✓ (4.67) | ✓ (4.67) | ✓ (4.83) | ✓ (5) | ✓ (5) | ✓ (5) |
| ollama/llama3 | — | — | — | — | — | — | — |
| openai/gpt-5.4 | ✓ (4.33) | ✓ (4.83) | ✓ (4.67) | ✓ (4.5) | ✓ (4.83) | ✓ (5) | ✓ (5) |
| openai/gpt-5.4-mini | ✓ (4.67) | ✓ (4.67) | ✓ (4.5) | ✓ (4.5) | ✓ (4.83) | ✓ (5) | ✗ (2.33)was ✓ (5) |
Model Status
- ↑ anthropic/claude-haiku-4-5 up
- → anthropic/claude-opus-4-6 stable
- → anthropic/claude-sonnet-4-6 stable
- ↑ gemini/gemini-2.5-flash up
- → gemini/gemini-2.5-pro stable
- → openai/gpt-5.4 stable
- ↓ openai/gpt-5.4-mini down
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