Unverified50% confidenceReleaseExact time
OpenAI has launched a premium speed tier called Ultrafast, which achieves up to 8x faster token generation in Codex and up to 6x faster speeds via the API.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/30/2026
First Seen
Valid until: 12/29/2026
Sources
Related Entities
Related Claims
UnverifiedOpenAI has committed to releasing a high-speed version with an inference speed of 750 tokens per second, approximately 10 times the speed of current mainstream competitors.76% similarUnverifiedUsing OpenCode to call DeepSeek is very fast, reaching over 100k tokens in approximately 3 to 5 minutes.68% similarUnverifiedUsing Codex in "fast" mode with "high" reasoning intensity, an e-commerce product selection assistant tool of 1,322 lines of code was generated in 3 minutes and 5 seconds.67% similarUnverifiedAn official demo showcasing Ultrafast with a racing game dynamically generating sprite images demonstrates that Ultrafast requests complete output first while regular generation tasks are still running66% similarUnverifiedQuickJS adopts a bytecode interpreter model; its execution speed is typically 1/10 to 1/5 that of V8, and its codebase is approximately 80,000 lines of C code.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/962544API
curl https://kongchang.com/api/v1/knowledge/claims/962544MCP
get_claim(id=962544)