Google launches Gemini 3.7 Flash for coding and agents, Pro model still delayed

Google has announced the availability of Gemini 3.7 Flash, its latest AI model in the Flash line, designed for coding and autonomous agent tasks. In an official announcement by Tulsee Doshi, Senior Director of Product Management on behalf of the Gemini team, the model was described as the most intelligent workhorse model the company has ever released. The launch price is set at $0.75 per million input tokens and $3.75 per million output tokens, positioning it as a competitive option in the mid tier AI model market. This launch comes at an unusual time for Google.

Bloomberg Technology reports that Alphabet Inc., Google's parent company, has not provided a release date for its most powerful AI model, Gemini 3.5 Pro, which has been hampered by repeated delays. The arrival of Gemini 3.7 Flash serves as a counterbalance amid uncertainty over the flagship model's release schedule. Gemini 3.7 Flash focuses on two primary areas: coding and agents. For coding, the model is designed to help developers write, review, and fix code more efficiently. For agents, it is intended to handle autonomous tasks that require planning and step by step execution, such as managing complex workflows or interacting with various applications independently. The pricing Google has set for Gemini 3.7 Flash places it in a more affordable segment compared to premium AI models.

At $0.75 per million input tokens and $3.75 per million output tokens, Google appears to be courting developers and companies that need high performance AI without paying the premium costs of top tier models. This pricing strategy also reflects the intense competition in the AI model market, where cost efficiency is a key adoption factor. The context of this launch cannot be separated from the turbulent dynamics of the AI industry. Google has long competed with OpenAI, Anthropic, and other players in the race to deliver the best AI models. The Flash line has traditionally been positioned as lighter and faster than the Pro models, yet still capable of handling demanding tasks like coding and data analysis. The delay of Gemini 3.5 Pro has drawn particular attention.

Bloomberg reports that the model has been held back by various obstacles, though specific details about the causes of the delay have not been disclosed. This situation raises questions about Google's readiness to deliver its most advanced AI model amid intensifying competitive pressure. From a market perspective, the launch of Gemini 3.7 Flash can be seen as a move by Google to maintain momentum in the segment of widely accessible AI models. With relatively affordable pricing, Google could attract a larger user base, especially startups and mid sized companies that need advanced AI capabilities without large budgets. However, it is worth noting that Google has not made an official statement about when Gemini 3.5 Pro will be released.

This lack of clarity creates uncertainty among enterprise users who may be waiting for the flagship model for more complex needs. In the meantime, Gemini 3.7 Flash offers a short term solution that can meet current coding and agent requirements. In its announcement, Google emphasized that Gemini 3.7 Flash is the most intelligent workhorse model it has ever built. The term "workhorse" refers to a model designed to handle everyday tasks with high reliability. This contrasts with premium models that are more geared toward experimental or advanced research tasks. The launch also demonstrates Google's strategy in managing its AI model portfolio. By offering a powerful Flash model, Google can serve a broader market while continuing to develop more advanced Pro models.

This approach allows Google to remain relevant without having to wait for the flagship release. On the technical side, Gemini 3.7 Flash is designed to handle long contexts and tasks that require multi step reasoning. This makes it suitable for agent applications that must plan and execute a series of actions autonomously. This capability is increasingly important as the trend toward AI agents is predicted to become a major growth area in the industry. The competitive token pricing also signals that Google is serious about competing in a cost sensitive AI model market. At $0.75 per million input tokens, Google offers a lower price than some premium models on the market, although direct comparisons with other models were not mentioned in the official announcement.

For developers, the availability of Gemini 3.7 Flash means they have a new option for building AI powered applications more cost effectively. This could drive AI adoption among those previously deterred by high costs, especially for applications requiring large scale processing such as log analysis, document automation, or coding assistants. Meanwhile, the delay of Gemini 3.5 Pro remains a major question mark. Bloomberg reports that the model has been hampered by delays, but provides no further details. This uncertainty could affect the confidence of some users waiting for Google's most advanced model, even though Gemini 3.7 Flash offers a solid alternative for current needs.

Going forward, Google faces the challenge of balancing the delivery of widely accessible models with the development of flagship models that push the boundaries of AI capability. The launch of Gemini 3.7 Flash shows that Google can still innovate in the workhorse segment, but the question of when Gemini 3.5 Pro will arrive remains unanswered. In the short term, Gemini 3.7 Flash will be Google's mainstay for attracting developers and companies needing AI for coding and agents. With competitive pricing and claimed best in class capabilities, this model has the potential to become a top choice for many development teams. However, Google's long term success in the AI market will heavily depend on its ability to release the long awaited Pro model without further delays.