Unverified50% confidenceFactExact time
The main experiments are based on Qwen3-1.7B and 4B configurations, with visual capacity fixed at 0.4B parameters, plus a scale ladder of backbone models from 0.6B to 6.4B.
1
Sources
50%
Confidence
Long-term
Relevance
11/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe study tested four sizes of the Qwen3 series—0.6B, 1.7B, 4B, and 8B—running at FP16 precision, greedy decoding, a single frozen prompt, and no fine-tuning.77% similarUnverifiedThe Qwen series of models is developed by Alibaba's Tongyi Lab, with 4B and 27B representing the parameter scales.76% similarUnverifiedThe main model of the architecture used by Qwen4 has 125B parameters, activates only 6B per token, uses a MoE architecture, contains 512 experts, and has a native context of 262K.74% similarUnverified7B–9B parameter models typically occupy 5–7 GB of VRAM after Q4 or Q5 quantization.72% similarUnverifiedIn testing, all three models—qwen2.5:3b, qwen2.5:1.5b, and llama3.2:1b—kept model memory below 0.5 GB, and all produced 0 hallucinated place names.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1173179API
curl https://kongchang.com/api/v1/knowledge/claims/1173179MCP
get_claim(id=1173179)