claude-haiku-4-5 lost common-sense-1; gemini-2.5-pro lost logic-1; gemini-2.5-pro lost math-1; gemini-2.5-pro lost spatial-1; gemini-2.5-pro lost causality-1; gemini-2.5-pro lost code-1; gemini-2.5-pro lost ambiguity-1; gemini-2.5-pro lost common-sense-1; gemini-2.5-flash lost logic-1; gemini-2.5-flash lost math-1; gemini-2.5-flash lost spatial-1; gemini-2.5-flash lost code-1; gemini-2.5-flash lost ambiguity-1; gemini-2.5-flash lost common-sense-1. gpt-5.4-mini dropped on spatial-1. gpt-5.4-mini failing spatial-1. gemini-2.5-flash recovering.
August 26, 2026 — 12:14 AM CT
Drift Alerts
- SCORE_DROP openai/gpt-5.4-mini on spatial-1
- REGRESSION anthropic/claude-haiku-4-5 on common-sense-1
- REGRESSION gemini/gemini-2.5-pro on logic-1
- REGRESSION gemini/gemini-2.5-pro on math-1
- REGRESSION gemini/gemini-2.5-pro on spatial-1
- REGRESSION gemini/gemini-2.5-pro on causality-1
- REGRESSION gemini/gemini-2.5-pro on code-1
- REGRESSION gemini/gemini-2.5-pro on ambiguity-1
- REGRESSION gemini/gemini-2.5-pro on common-sense-1
- REGRESSION gemini/gemini-2.5-flash on logic-1
- REGRESSION gemini/gemini-2.5-flash on math-1
- REGRESSION gemini/gemini-2.5-flash on spatial-1
- IMPROVEMENT gemini/gemini-2.5-flash on causality-1
- REGRESSION gemini/gemini-2.5-flash on code-1
- REGRESSION gemini/gemini-2.5-flash on ambiguity-1
- REGRESSION gemini/gemini-2.5-flash on common-sense-1
Provider Status
- Anthropic Issues logging into Claude.ai
- Anthropic Errors logging into Claude.ai
- Anthropic Elevated errors for multiple models
Scorecard
| Model | ambiguity-1 | causality-1 | code-1 | common-sense-1 | logic-1 | math-1 | spatial-1 |
|---|---|---|---|---|---|---|---|
| anthropic/claude-haiku-4-5 | ✓ (4.5) | ✓ (4.5) | ✓ (5) | ✗ (3)was ✓ (3.33) | ✓ (5) | ✓ (5) | ✓ (5) |
| anthropic/claude-opus-4-6 | ✓ (5) | ✓ (4.75) | ✓ (5) | ✓ (4.5) | ✓ (5) | ✓ (5) | ✓ (5) |
| anthropic/claude-sonnet-4-6 | ✓ (4.5) | ✓ (5) | ✓ (5) | ✓ (4.25) | ✓ (5) | ✓ (5) | ✓ (5) |
| gemini/gemini-2.5-flash | — | — | — | — | — | — | — |
| gemini/gemini-2.5-pro | — | — | — | — | — | — | — |
| ollama/llama3 | — | — | — | — | — | — | — |
| openai/gpt-5.4 | ✓ (4.5) | ✓ (4.75) | ✓ (5) | ✓ (4.75) | ✓ (4.5) | ✓ (4.5) | ✓ (5) |
| openai/gpt-5.4-mini | ✓ (4.75) | ✓ (4.75) | ✓ (4.75) | ✓ (4.75) | ✓ (5) | ✓ (5) | ✗ (2)was 3.5 |
Model Status
- ↓ anthropic/claude-haiku-4-5 down
- → anthropic/claude-opus-4-6 stable
- → anthropic/claude-sonnet-4-6 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