[KongchangAI]
Unverified50% confidenceEventExact time

William Liu compared the tokenization results of multiple GPT models using a set of fixtures, with the experiment covering seven GPT models, including GPT-5.5, three variants of GPT-5.6 (Sol/Terra/Luna), and three variants of GPT-6 (Astra/Sol/Luna).

1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027

Sources

ttok 1.0 Released: Default Tokenizer Switches to GPT-5/GPT-6

Related Entities

William LiupersonGPT-5.5modelGPT-5.6modelGPT-6model

Related Claims

UnverifiedGPT-5.6 Codex评测中设置了Sol、Terra、Luna三个不同的模型档位配置78% similarUnverifiedAccording to Artificial Analysis data, GPT-6-Sol and 5.6-Sol are approximately 100 tok/s, while GPT-6.1-Sol and GPT-6-Astra are approximately 60 tok/s.75% similar

Cite This Claim

Stable URI
https://kongchang.com/claim/1009725
API
curl https://kongchang.com/api/v1/knowledge/claims/1009725
MCP
get_claim(id=1009725)