Batch requests are now available on Merge Gateway!
Not every request needs an answer immediately.
Use our Batch API to route non-latency sensitive tasks for significantly cheaper prices. The same results at a fraction of the cost.
Show more
The model that tops agentic tool calling costs $0.50 per million output tokens.
GLM-5.3-Flash leads Toolathlon at 78.4%, and five of the top six run under $5 per million out. Price and tool-calling accuracy have come apart at the top of this board.
That's an argument for routing on the benchmark instead of committing to one model up front. Weight Toolathlon in a Build-Your-Own-Router policy and Gateway sends each tool-heavy request to whatever leads it that week.
Scores via
Show more
We're at CISO New York 2026!
Stop by our booth for some swag and the chance to win a Mac Mini.
We're partnering with Nebius (@NebiusAI), Tavily (@Tavily), and Replit (@Replit) for Builders & Brews SF on October 9, 3 - 9 PM.
RSVP here:
One week left on 90% off GLM 5.3 Flash.
A task costs nine tenths of a cent until September 30th, then nine cents. If you have a batch job, an eval sweep or a test suite you have been meaning to run, this is the week.
Show more
Gemini 3.8 Flash TTS and Flash-Lite TTS from Google (@Google) are now available on Merge Gateway.
They take the top two spots on Hume AI's Overall Quality Index, and Flash TTS leads the Voice Design Benchmark at 71.4.
Show more
They Bay's been moving different since we dropped these.
MCP getting spicy on main
GPT-6 Sol and GPT-6 Luna from OpenAI (@OpenAI) are now available on Merge Gateway.
Both cost half as much as the 5.6 series, and OpenAI says Sol makes about half as many mistakes as GPT-5.6 Sol.
Sol for coding, Luna for high-volume extraction and summaries.
Check it out:
Show more
.@AnthropicAI just shipped Claude Opus 5.5!
On most tasks it holds its own against Fable 5.1, and it runs at 40% less cost than Opus 5.
Check it out:
Seven cents per million output tokens. DeepSeek charges $0.66 for the same model.
@deepseek_ai V4 Flash is 75% off in Merge Gateway until September 30. Output is where a long agent run actually spends, so it is the number worth checking before the discount ends.
Show more
We're partnering with Nebius (@NebiusAI), Tavily (@Tavily), and Replit (@Replit) for Builders & Brews SF, a stop on their 20-city cafe tour for AI builders.
Join us October 9, 3-9 PM - RSVP here:
Show more
Browserbase (@browserbase) is now available in Merge Agent Handler.
Agents can fetch a page, search the web, or hand a plain-English task to a browser agent that runs on its own.
Check it out:
Show more
Grok 4.7 brings frontier-level knowledge work at a fraction of Fable 5.1’s token cost.
You're literally getting 80% lower input pricing and 88% lower output pricing for a minuscule gap in performance.
Great work @SpaceXAI and @elonmusk!
Show more
Grok 4.7 from xAI (@SpaceXAIx) is now available on Merge Gateway.
Coding is the story. It scores 46.3% on CursorBench 4.0 against 40.4% for Grok 4.6, and 71.0% on DeepSWE v1.1.
Check it out:
Show more
Jev from TypeSafe AI (@typesafeai) is now available on Merge Gateway.
Send unstructured state and a typed question, and Jev returns a typed decision with a calibrated probability with little latency.
Input runs $0.042 per million tokens and output is free.
Show more
Your agents have new tools. Four new connectors just landed in Merge Agent Handler:
- Ordinal (@tryordinal)
- Browserbase (@browserbase)
- Discord (@discord)
- Copper (@CopperInc)
Start building:
Show more
“Less of the mundane, more of the hard. Product Managers can focus on solving really hard problems.”
Last week we sat down with some of the best in product to talk about how the role has evolved in the age of AI. How are Product Managers adapting and what were their takeaways from Lenny & Friends.
Special thanks to Rodrigo Davies, Christina Ferancik, Lucas Dickey, Julia Kanter, Dante Alvarado-Leon, Vanisha Wason, Matthew Himelstein
Show more
GPT-6 Astra leads Terminal-Bench 4.0 among priced models at 57.7%, with Claude Fable 5.1 close behind at 55.8%.
Claude Opus 5 reaches 51.8% at half the output price.
With Merge Gateway, you can route terminal tasks to the model that fits the performance and cost target you set.
Show more
