IMPLICATOR
Google's Gemini rose to 90 in Implicator's weekly LLM Meter for the week of May 24, extending the lead it took from Anthropic's Claude a week earlier. The gain of two points followed Gemini 3.5 Flash, which Google released at its I/O developer conference on May 19. The meter scores six large language models each week from the perspective of enterprise and business buyers.
Gemini 3.5 Flash scored 55 on the independent Artificial Analysis Intelligence Index, within two points of Claude Opus 4.7 and five of OpenAI's GPT-5.5, at roughly a third of the per-token cost, according to Google's launch post. Google named Salesforce, Databricks, Ramp, and Xero as launch customers, with Salesforce folding the model into its Agentforce platform. The company also upgraded its Antigravity development platform and introduced Google Spark, a cloud-based agent that works tasks in the background.
Key Takeaways
- Gemini rose to 90 in Implicator's weekly LLM Meter, extending its lead after Google I/O on May 19.
- Gemini 3.5 Flash scored within two points of Claude Opus 4.7 at roughly a third of the per-token cost.
- Claude slipped to 86 on price pressure and a June 15 billing change; ChatGPT rose to 85 on enterprise deals.
- Mistral, Grok, and DeepSeek trail at 72, 31, and 16; DeepSeek cut its flagship API price by 75%.
AI-generated summary, reviewed by an editor. More on our AI guidelines.
What the launch settled
The result resolved a risk the meter flagged the prior week, when preview coverage suggested the new Gemini might ship behind GPT-5.5. A lower-cost model instead landed close to the field's reference flagship. Google said its larger Gemini 3.5 Pro is in internal use and will release next month, so Claude Opus 4.7 still holds the top of the quality field for now. A DeepMind employee letter over classified Pentagon work remains unresolved.
Claude slips to 86
Claude fell one point to 86. Opus 4.7 remains the model rivals benchmark against, and Anthropic's enterprise stack from earlier in May, including its PwC and SAP partnerships, stayed in place. The downgrade reflected price pressure from Gemini 3.5 Flash and a billing change that takes effect June 15, when Anthropic moves programmatic Claude usage to a separate metered credit pool. The company has not published the token sizes behind its subscription caps.
Track the AI model race every morning
Strategic AI news from San Francisco. No hype, no "AI will change everything" throat clearing. Just what moved, who won, and why it matters. Daily at 6am PST.
No spam. Unsubscribe anytime.
ChatGPT gains on enterprise deals
ChatGPT rose one point to 85. OpenAI and Dell Technologies said on May 18 they would bring the Codex coding agent to hybrid and on-premises environments through the Dell AI Data Platform, a deployment option for buyers that cannot send code to a public endpoint. Gartner named OpenAI a Leader in enterprise coding agents on May 22, and the company committed more than $234 million to a new applied AI lab in Singapore, its first outside the United States.
Lower in the field
Mistral fell one point to 72 after a quiet week whose main event was its purchase of Emmi AI, a Vienna startup building physics-aware models for industrial engineering. Grok rose one point to 31. xAI shipped Grok Skills, third-party connectors, and a coding agent during the week, but a SpaceX filing tied to its planned public offering disclosed that xAI spent $6.4 billion last year, alongside another round of layoffs. DeepSeek rose one point to 16 after cutting the API price of its V4-Pro model by 75%, to about $0.44 per million input tokens, while its first external funding round at a $45 billion to $50 billion valuation continued. US government-device bans on the Chinese model remain in force.
The next meter publishes May 31. Gemini 3.5 Pro is scheduled to ship before then.
Frequently Asked Questions
What is the Implicator LLM Meter?
It is Implicator's weekly scorecard rating six large language models from an enterprise-buyer perspective, weighing compliance, model quality, reliability, ecosystem maturity, vendor stability, and pricing. Scores run from 0 to 100 and update each week.
Why did Gemini rise this week?
Google released Gemini 3.5 Flash at its I/O conference on May 19. The model scored within two points of Claude Opus 4.7 on the Artificial Analysis index at roughly a third of the cost, with Salesforce, Databricks, Ramp, and Xero named as launch customers.
Why did Claude slip?
Claude fell one point to 86 on price pressure from Gemini 3.5 Flash and a June 15 billing change that moves programmatic Claude usage to a separate metered credit pool. Claude Opus 4.7 still holds the top of the quality field.
What helped ChatGPT?
OpenAI and Dell agreed to bring the Codex coding agent to hybrid and on-premises environments, Gartner named OpenAI a Leader in enterprise coding agents, and OpenAI committed more than $234 million to a new applied AI lab in Singapore.
Where do Mistral, Grok, and DeepSeek stand?
Mistral fell to 72 after a quiet week, Grok rose to 31 despite a $6.4 billion annual spend disclosed in a SpaceX filing, and DeepSeek rose to 16 after cutting its V4-Pro API price 75%. US government-device bans on DeepSeek remain in force.
AI-generated summary, reviewed by an editor. More on our AI guidelines.
Marcus Schuler
Editor-in-Chief and founder of Implicator.ai. Former ARD correspondent and senior broadcast journalist with 10+ years covering tech. Writes daily briefings on policy and market developments. Based in San Francisco. E-mail: [email protected]
