Unverified50% confidenceFactExact time
The specific composition and cleaning methods of Hydrogen-1.0's training data are not disclosed, which affects outside judgment of the model's capability boundaries.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/2/2026
First Seen
Valid until: 1/31/2027
Sources
Related Entities
Related Claims
UnverifiedHydrogen-1.0 was not pretrained from scratch but post-trained on Z.ai's GLM-5.3.75% similarUnverified42AI has not disclosed the model's specific architecture, training data, or performance.62% similarUnverifiedPrivately held benchmark test data is not publicly available, meaning models cannot have seen the evaluation set during training, which more accurately reflects generalization ability and avoids leaderboard contamination.61% similarUnverifiedModels trained solely on external data cannot reliably generalize to the L2-ARCTIC dataset, indicating significant domain mismatch between different corpora.58% similarUnverifiedThe original article did not disclose joey-1's training data composition, calibration method or evaluation benchmarks.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1128056API
curl https://kongchang.com/api/v1/knowledge/claims/1128056MCP
get_claim(id=1128056)