Our content generation service is experiencing issues. A human-curated summary is being prepared.
LLMs & Generative AI

AI tools enable product search, store calls and auto‑buy; Baidu unveils ERNIE 5

2 min read

The past week has been a busy one for large‑language‑model developers. Gemini 3 rolled out with a tighter dialogue loop, Opus 4.5 pushed its multimodal rendering, Nano Banana Pro claimed a new efficiency record, and GPT‑5.1‑Codex‑Max hinted at deeper code‑generation tricks. Those releases sit alongside a quieter but potentially more consumer‑facing shift: companies are turning chat‑style AI into a shopping assistant.

While the hype often focuses on headline‑grabbing models, the real question is how quickly those models move from research labs into everyday transactions. Baidu’s recent showcase at Baidu World 2025 put that question front and center, unveiling its next‑generation ERNIE 5.0 and a suite of applications aimed at the retail aisle. The event signaled a broader push to embed conversational agents in the buying process, from checking inventory to sealing a deal when a price drops.

In that context, the following details illustrate exactly what’s being tested on the ground.

New tools enable conversational product searches, let an AI call local stores for stock and deals, and authorize an AI to automatically buy items when prices hit a set threshold. Baidu Unveils ERNIE 5.0 and a Series of AI Applications at Baidu World 2025, Ramps Up Global Push. The company showcased upgrades across its AI portfolio--including the natively omni-modal ERNIE 5.0, new and improved digital human and agent products like Famou and GenFlow 3.0, global rollouts for tools such as MeDo and Oreate--and reported Apollo Go has completed over 17 million driverless rides.

Related Topics: #AI #LLM #Gemini 3 #Opus 4.5 #GPT‑5.1‑Codex‑Max #ERNIE 5.0 #Baidu World 2025 #Apollo Go #GenFlow 3.0

Did the latest releases finally delivered on hype? Google’s Gemini 3 entered the Gemini app and AI Search this week, boasting record benchmark scores and a new coding app. A research‑tier Gemini 3 Deepthink is slated for AI Ultra subscribers, but only after further safety testing, leaving its broader availability uncertain.

Meanwhile, new AI tools now claim to conduct conversational product searches, place calls to local stores for stock and deals, and even authorize purchases when prices dip below a set threshold. Baidu announced ERNIE 5.0 at Baidu World 2025, alongside a suite of applications and a push toward global markets. The company showcased the model’s capabilities, yet details on performance or integration remain sparse.

Nvidia’s earnings were also highlighted, though the impact on the AI hardware supply chain was not elaborated. Collectively, the announcements suggest rapid feature expansion, but it is unclear whether developers and consumers will adopt these functions at scale, or how safety safeguards will evolve. The week underscored both progress and lingering questions about practical rollout.

Further Reading

Common Questions Answered

What new shopping‑assistant capabilities do the AI tools mentioned in the article provide?

The article says the latest AI tools can conduct conversational product searches, place phone calls to local stores to check stock and deals, and even authorize automatic purchases when prices fall below a preset threshold. These functions turn chat‑style models into end‑to‑end shopping assistants.

Which model did Baidu unveil at Baidu World 2025 and what are its key features?

Baidu introduced the natively omni‑modal ERNIE 5.0, which supports seamless integration of text, image, and audio inputs. The rollout also included upgraded digital‑human agents such as Famou and GenFlow 3.0, highlighting a broader global push for AI applications.

How does Gemini 3 differ from earlier Gemini releases according to the article?

Gemini 3 arrives with a tighter dialogue loop and has been integrated into the Gemini app and AI Search, achieving record benchmark scores. A research‑tier variant called Gemini 3 Deepthink is slated for AI Ultra subscribers after additional safety testing.

What efficiency claim does Nano Banana Pro make in the article?

Nano Banana Pro is reported to have set a new efficiency record among large‑language‑model developers, indicating it delivers higher performance per compute unit than its predecessors. This claim positions it alongside other headline‑grabbing releases like Opus 4.5 and GPT‑5.1‑Codex‑Max.