ITmedia compared three AI chatbots' speed and stability. ChatGPT was fastest to start answering.
Gemini was most stable across day and evening.
Claude Sonnet 5 was slowest to complete responses.
What happened
ITmedia's ICT Research Division tested ChatGPT, Gemini, and Claude Sonnet 5 in July 2026. ChatGPT had the fastest median time to first token at 0.70 seconds (day) and 0.73 seconds (evening). Claude Sonnet 5 was slowest overall with 3.28–3.41 seconds end-to-end.
Why it matters
Speed and stability vary by time of day and service. Gemini showed the least variation (standard deviation 0.20 seconds), making it the most consistent. Claude Sonnet 5's evening variability was larger, possibly due to Japanese users overlapping with US/Europe peak hours.
What to watch
For businesses relying on API-based AI, these differences affect user experience. The test used specific models (gpt-5-chat-latest, gemini-3.5-flash, claude-sonnet-5) over a wired LAN without VPN, so real-world results may differ.
Ask the AI about this article →
The benchmark, conducted by ITmedia's ICT Research Division on July 14–15, 2026, measured API-based response times for three major AI chatbots. The test used specific model versions and sent ten short prompts over two time slots, with output limited to 200 tokens. Results show a clear trade-off: ChatGPT leads in speed to first token, while Gemini excels in stability, and Claude Sonnet 5 trails in end-to-end latency.
The findings highlight that AI performance is not constant. Variations between day and evening suggest infrastructure load matters, especially for global services. For businesses integrating these APIs, choosing a model involves balancing speed against consistency, depending on user expectations and peak usage times.
Since the test used a wired LAN without VPN, real-world conditions may differ. The article notes that individual results can vary based on network and server conditions, so these numbers serve as a reference point rather than a definitive ranking.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
Nvidia is reportedly in discussions to invest in AI startup Perplexity as part of a funding round that would v…
SK Hynix is pushing beyond HBM (high-bandwidth memory, a key component in AI chips) to two new areas: HBF and…
dotData explained its approach to using a separate LLM (an AI that understands and generates text) to rate ano…
Anthropic's Claude AI service is currently experiencing an outage, as indicated by the article title
Taiwan prosecutors indicted nine people, including employees of Nvidia and Super Micro, for allegedly exportin…
Snowflake introduced dynamic model routing in Cortex AI Gateway, which automatically selects the most affordab…
