Unverified50% confidenceFactExact time
Timestral offers models in four sizes, with the largest at 209 million parameters and the smallest at 5 million parameters.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/2/2026
First Seen
Valid until: 1/31/2027
Sources
Related Entities
Related Claims
VerifiedThe GPU resources required for a single inference of a hundred-billion-parameter flagship model may be 10 to 20 times higher than those of a ten-billion-parameter model.68% similarUnverifiedA micro CNN may have only thousands to tens of thousands of parameters, with a single inference computation on the order of 1–10 MFLOPs and an inference latency in the millisecond or even sub-millisecond range.68% similarUnverifiedBefore the era of large models, the largest model the presenter had trained had only approximately 4,000 parameters.68% similarUnverifiedThe Timestral-209M model reads up to 32,768 historical steps, predicts up to 1,024 steps per forward pass, and outputs 21 quantiles (0.01 to 0.99) per prediction.68% similarUnverifiedThe 4-billion-parameter version of Qwen 3.5, with no external tools and with exact calculation required at each step, got only 22 of 40 multi-digit multiplication problems correct.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1128091API
curl https://kongchang.com/api/v1/knowledge/claims/1128091MCP
get_claim(id=1128091)