Unverified50% confidenceFactExact time
Gemma 4 26B A4B has 25.2B total parameters / 3.8B active; Qwen3-Next-80B-A3B has 80B / 3B; gpt-oss-20b has 20.9B / 3.6B.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedGemma 4 26B-A4B ranked first and was described as an upgrade over Qwen 35B-A3B, with better dialogue and monologue and stronger instruction-following.78% similarUnverifiedGemma 4 26B-A4B is a MoE model that activates approximately 4B parameters per token, with 26B being the total parameter count and A4B representing approximately 4 billion parameters activated per token.75% similarUnverifiedThe 3.8-27B has fewer parameters than the 3.5/3.6-35B-A3B series.73% similarUnverifiedGemma-3 was released in early 2025 and supports different parameter scales ranging from 2B to 27B.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1003833API
curl https://kongchang.com/api/v1/knowledge/claims/1003833MCP
get_claim(id=1003833)