Unverified50% confidenceFactExact time
Reflection stated that Beam achieves comparable scores to GLM on reasoning benchmarks while consuming three to four times less inference compute, a figure estimated based on activated parameters and generated token counts rather than measured costs.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/8/2026
First Seen
Valid until: 1/6/2027
Sources
Related Entities
Related Claims
VerifiedThe Mixture of Experts (MoE) architecture activates only approximately 10%–30% of total parameters during inference, significantly reducing FLOPs and GPU memory usage per inference63% similarUnverifiedThe actual compute of a single GLM 5.3 inference may be equivalent to that of a dense model with 70B–150B parameters.62% similarVerifiedA 2023 Stanford University study showed that multimodal models such as GPT-4V, when parsing flowcharts with nested conditional logic, had accuracy approximately 15–25 percentage points lower than equivalent textual descriptions.62% similarUnverified对同一测试用例执行多次调用统计成功比例得到通过率，比二元的通过/失败判断更能反映模型真实表现62% similarUnverifiedEach token in the GLM series 744B model activates only approximately 40B parameters, accounting for about 5.4% of the total.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/987293API
curl https://kongchang.com/api/v1/knowledge/claims/987293MCP
get_claim(id=987293)