Moonshot Launches Kimi 2.7 Code High-Speed Model, Boosting Output Speed Sixfold
Moonshot AI followed last week's launch of the Kimi 2.7 Code model with a high-speed variant released tonight. This version delivers a significant boost in output speed while preserving the core capabilities of the original, aiming to help developers write code more efficiently.

High Speed Dramatically Reduces Waiting Time
According to official data, the high-speed Kimi 2.7 Code model outputs 5 to 6 times faster than the standard version. In typical coding scenarios, it reaches around 180 tokens per second, while in short-context tasks the maximum speed can hit 260 tokens per second.
On pricing, the high-speed API costs only twice as much as the regular version, delivering several times the efficiency at a modest premium. Kimi Code Plan users can join the early access program to experience this in Kimi Code first.
Thinking Mode Unlocks Full Performance
To ensure the model performs at its best, users must enable the thinking mode when using Kimi 2.7 Code series models. The Kimi API and Kimi Code have this feature turned on by default. If users disable it manually, the API will throw an error and the system will automatically revert to the K2.6 version.
As computing power continues to expand, Moonshot AI will broaden access to this high-speed model. Starting July this year, it will gradually open to members at the Allegretto level and above. At that time, the usage cost in Kimi Code Plan will be adjusted to three times the regular version.
Related article
Stripe to Acquire OpenRouter as AI Model Routing Gains Momentum
Stripe has agreed to acquire OpenRouter, an AI model-routing platform that gives developers access to hundreds of models through a single interface. The deal adds model selection and routing to Stripe’s existing work around AI usage and token-based b
Former OpenAI Researcher Barrett Zoph Returns to Google to Support Gemini Development
Barret Zoph, who previously abruptly left and joined OpenAI due to a disagreement with Thinking Machines Lab CEO Miriam Mularati, has once again changed employers. Alphabet's Google announced that Zoph will return as Vice President of Research, bring
How Recursive Intelligence Raised $335M at $4B Valuation in 4 Months
The co-founders of startup Ricursive Intelligence seemed destined to be co-founders.Anna Goldie, CEO, and Azalia Mirhoseini, CTO, are so well-known in the AI community that they were among those AI engineers who “got those weird emails from Zuckerber
Related Special Topic Recommendations
Comments (1)
0/500
Moonshot AI followed last week's launch of the Kimi 2.7 Code model with a high-speed variant released tonight. This version delivers a significant boost in output speed while preserving the core capabilities of the original, aiming to help developers write code more efficiently.

High Speed Dramatically Reduces Waiting Time
According to official data, the high-speed Kimi 2.7 Code model outputs 5 to 6 times faster than the standard version. In typical coding scenarios, it reaches around 180 tokens per second, while in short-context tasks the maximum speed can hit 260 tokens per second.
On pricing, the high-speed API costs only twice as much as the regular version, delivering several times the efficiency at a modest premium. Kimi Code Plan users can join the early access program to experience this in Kimi Code first.
Thinking Mode Unlocks Full Performance
To ensure the model performs at its best, users must enable the thinking mode when using Kimi 2.7 Code series models. The Kimi API and Kimi Code have this feature turned on by default. If users disable it manually, the API will throw an error and the system will automatically revert to the K2.6 version.
As computing power continues to expand, Moonshot AI will broaden access to this high-speed model. Starting July this year, it will gradually open to members at the Allegretto level and above. At that time, the usage cost in Kimi Code Plan will be adjusted to three times the regular version.
Stripe to Acquire OpenRouter as AI Model Routing Gains Momentum
Stripe has agreed to acquire OpenRouter, an AI model-routing platform that gives developers access to hundreds of models through a single interface. The deal adds model selection and routing to Stripe’s existing work around AI usage and token-based b
Former OpenAI Researcher Barrett Zoph Returns to Google to Support Gemini Development
Barret Zoph, who previously abruptly left and joined OpenAI due to a disagreement with Thinking Machines Lab CEO Miriam Mularati, has once again changed employers. Alphabet's Google announced that Zoph will return as Vice President of Research, bring
How Recursive Intelligence Raised $335M at $4B Valuation in 4 Months
The co-founders of startup Ricursive Intelligence seemed destined to be co-founders.Anna Goldie, CEO, and Azalia Mirhoseini, CTO, are so well-known in the AI community that they were among those AI engineers who “got those weird emails from Zuckerber





Home






