Expired50% confidenceBenchmarkExact time
Gemma 4 26B-A4B (thinking mode) scored only 0.07; with thinking disabled it never edited files across the 9 tasks, and with it enabled it would edit but was mostly wrong.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/2/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 11/16/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedZero-shot Gemma 3 270M is nearly unusable on the grammar correction task, with an F0.5 of only 0.122, and 99.6% of correct sentences were altered.73% similarUnverifiedGemma 4 E2B comprehensively surpasses Gemma 3 27B on reasoning, coding and agentic tasks; E4B consistently beats Gemma 3 27B on every benchmark.72% similarUnverifiedGemma 4 has thinking mode enabled by default in Ollama: with think=True, 58 seconds, 248 tokens, wrong answer; with think=False, 5 seconds, 9 tokens, correct answer.71% similarUnverifiedThe Gemma 4 31B version, while maintaining a relatively small parameter count, attempts to strike a balance between conversational quality and breadth of knowledge.71% similarUnverifiedThe author empirically compared Gemma 3 1B and 4B: 1B, in the rename scenario, wrote "the app will resend the shelved event" and told the sender to make the change, and both errors were caught by the checks and fell back to the rule answer; 4B took about 16 seconds on a laptop GPU and accurately stated the customer, amount and specific field.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1127039API
curl https://kongchang.com/api/v1/knowledge/claims/1127039MCP
get_claim(id=1127039)