Unverified50% confidencePredictionExact time
The full v1 will continue pretraining to approximately 10 billion tokens and will incorporate training on long agent sessions to narrow the gap on agent tasks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/5/2026
First Seen
Valid until: 1/3/2027
Sources
Related Entities
Related Claims
UnverifiedAgent tasks require multiple model invocations, and a 12B model on a consumer-grade GPU already takes several seconds per inference, causing the time cost of multi-step tasks to accumulate significantly.66% similarVerifiedAn Agent's token consumption for a single task may be 5–10 times that of an ordinary conversation, due to the involvement of multiple rounds of large model calls.65% similarUnverifiedAn Agent handling a medium-scale code refactoring task may need to read 20–30 files, execute more than 10 tool calls, and accumulate a context exceeding 50,000 tokens.63% similarUnverifiedCLM's 8-billion-parameter backbone Qwen3-8B is frozen and serves only as a reader; what is actually trained are only the small projection heads on each side with approximately 20 million parameters each, and a complete pretraining run takes about one hour on a single RTX.63% similarUnverifiedUnitree has open-sourced the humanoid robot foundation model Unifor LMW-LA 1.0, with a parameter scale of 6 billion, trained on approximately 2,500 hours of real robot data; a single model can perform 64 tasks including getting out of bed, folding clothes, and sorting, and is compatible with multiple types of robotic hands.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/977119API
curl https://kongchang.com/api/v1/knowledge/claims/977119MCP
get_claim(id=977119)