Unverified50% confidenceBenchmarkExact time
The same parameter name reasoning_effort behaves differently across models from different vendors: in gpt-5.4-mini the median output tokens rose from 18 at none to 254 at high (×14.1); gpt-oss-120b rose from refusal to 1,340 (×4.6); claude-haiku-5.5 rose from 150 to 298 (×2.0); gemini-3.7-flash rose from 453 to 1,013 (×2.2).
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/12/2026
First Seen
Valid until: 10/26/2026
Sources
Related Entities
Related Claims
UnverifiedIn the hard-to-verify category with injection bait, the number of questions for which GPT-5.4 mini judged incorrectly in both orderings rose from 3 to 39, Qwen3-Next 80B Instruct from 4 to 38, and GPT-5.4 mini and Qwen3-Next Instruct scored zero on hard questions (below the approximately 0.25 expected from random coin flipping).73% similarUnverifiedThe open-source model MiniMax M2.6 Pro scored 46 points, while GLM 5.3 and Qwen 3.8 Max scored approximately 45 points, both close to GPT-6 Sol's score of 48 points.70% similarUnverifiedArtificial Analysis noted that Haiku 5.5 used 162,000 output tokens in the intelligence index test in Max mode, approximately three times that of GPT-6 Luna.70% similarUnverifiedGPT-3 reached a parameter scale of hundreds of billions (175 billion parameters) and demonstrated few-shot and zero-shot capabilities.69% similarUnverifiedIn the PR split-review example, GPT-6.1 Sol consumed input=77, output=266, and reasoning=149 tokens, took 5.4 seconds, did not fabricate diffs, and did not claim to have run tests.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1007693API
curl https://kongchang.com/api/v1/knowledge/claims/1007693MCP
get_claim(id=1007693)