Unverified50% confidenceBenchmarkExact time
The 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.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/18/2026
First Seen
Valid until: 1/16/2027
Sources
Related Entities
Related Claims
UnverifiedGemma 4 26B terminated with a runtime error before producing a score, because the scoring code assumed at least one sentence would be returned in the correct structured format.72% similarUnverifiedIn terms of inference speed, Gemma 4 E4B takes approximately 0.9 seconds per photo and E2B approximately 0.6 seconds, completing 165 photos within three minutes.70% similarUnverifiedGemma 4采用混合专家架构（MoE），E2B后缀表示Effective 2B，每次推理只激活约20亿有效参数68% similarUnverifiedAfter changing the header identification to have Gemma 4 E2B fill only the three required gaps of date, quantity, and product, all 12 header sets were correct with 0 errors, but the author cautions that the same batch of data was also used for debugging, so 12/12 should be regarded as an optimistic estimate.67% 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.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1061378API
curl https://kongchang.com/api/v1/knowledge/claims/1061378MCP
get_claim(id=1061378)