← Back to People
Confirmed that Azure's 80% benchmark score was caused by running an older vLLM version, now fixed.
1
Mentions
1
Articles
1
Outlets
Associated Topics
Companies Covered
Coverage Patterns
How media typically covers Lucas Pickup
Article Types
- analysis1
Media Angles
- technical1
Narrative Framing
- trend1
Associated AI Models
gpt-oss-120b1
Articles
Most recent firstQuoted In
Directly quoted in these articles
Open Weight LLMs Exhibit Inconsistent Performance Across Providers
Simon WillisonanalysiscautiousAug 18, 2025
The same open weight LLM (gpt-oss-120b) achieves dramatically different performance across hosted providers (36.7% to 93.3% on AIME), due to differences in serving frameworks, quantization, and configuration versions.
“Confirmed that Azure's 80% benchmark score was caused by running an older vLLM version, now fixed.”
Open Source AILLMsAI InfrastructureFoundation Models