ModelsPublished Top storyAll news from that day
Ongoing case: Frontier model API prices New model · 21 storiesGoogle releases Gemini 4 Neon for complex coding, analysis and enterprise workflows
Google says Neon ranks third on the Artificial Analysis Intelligence Index at its xhigh setting; it costs $3 per million input tokens and $15 per million output tokens.
Check our sources · 17 facts from 2 sources
Image: Google
Key points
- Google says Neon is built for complex coding, analytical reasoning and production-grade enterprise workflows, between Carbon for the hardest long-horizon projects and Hydrogen for fast everyday work.
- Google reports Neon at 61 on the Artificial Analysis Intelligence Index at xhigh (third overall), 58.6 on its Coding Agent Index (first) and 65.2% on Terminal-Bench 4.1.
- Neon costs $3 per million input tokens and $15 per million output tokens, with a 1 million-token context window, and every token the model processes is billed, thinking tokens included.
What happened
Google released Gemini 4 Neon, a model it says is built for complex software engineering, knowledge work and science.
Google's developer guide pitches it for complex coding, analytical reasoning and production-grade enterprise workflows, between Gemini 4 Carbon for the hardest long-horizon projects and Gemini 4 Hydrogen for fast everyday work.
The details are from Google's launch post of 18 November and its developer guide, which was updated on 19 November.
The performance claims are Google's own, reported by Google from benchmarks run by others. Google says Neon ranks third overall with a score of 61 on the Artificial Analysis Intelligence Index at its xhigh thinking level, and first on that firm's Coding Agent Index with 58.6.
It says Neon scores 65.2% on Terminal-Bench 4.1, 12.8% on SWE-sweep with the mini-SWE harness against 39.0% inside Google's Antigravity coding tool, and 46.2% on FrontierCode 2.0 at its default high setting.
Cost per task is the other claim. Google says that at the default high setting Neon pairs a high Artificial Analysis Intelligence Index score with a 32% lower cost per task than Carbon, and that at xhigh it costs 36% or more less per task than competing models at their highest settings. Neon has four thinking levels, LOW, MEDIUM, HIGH and XHIGH, and HIGH is the default.
Neon is rolling out in the Gemini API in Google AI Studio, in Gemini for business, in Antigravity for paid Google AI Pro and Ultra subscribers, and in the Gemini app for Pro and Ultra subscribers. On Google's enterprise platform the developer guide says it is available to verified enterprise customers in select regions.
Google says a Neon version for cyber defence is coming to trusted defenders in its Fairwind Program within days.
Google prices Neon at $3 per million input tokens, $0.15 per million cached input tokens and $15 per million output tokens, with a context window of 1 million tokens. The developer guide sets one limit of 1,048,576 tokens for input and output together, where output includes thinking, and says every token the model processes is billed, thinking tokens included.
What it means for you
Our viewTreat the ranks and scores as Google's claims until an independent run repeats them. The cost claims also come from Google: at its default high setting Neon costs 32% less per task than Carbon, and at xhigh 36% or more less than competing models at their highest settings.
Because thinking tokens are billed, the level you pick affects the bill as well as the result. Enterprise access is limited to verified customers in select regions. Run a sample of your own coding tasks at high and at xhigh, record the cost per task at each, and compare that with the model you use today before you switch.
It adds no new facts.
Your reaction
We count reactions per story and day, never who reacted. The counts help us choose what goes in the monthly issue. If you are signed in, your own page shows yours too.
We count reactions per story and day, never who reacted. They help choose what goes in the monthly issue.
Check our sources
Every sentence above is checked against these 2 sources.
1 Introducing Gemini 4 Neon
Open the source archived copy-
Google's launch post of 18 November 2026 says Gemini 4 Neon brings frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work and science workflows.
citeIntroducing Gemini 4 Neon Nov 18, 2026 ... Today, we’re introducing Gemini 4 Neon — bringing frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work, and science workflows.
-
Google says Gemini 4 Neon ranks #3 overall with a score of 61 on the Artificial Analysis Intelligence Index at its xhigh thinking level, landing just behind larger, slower models.
citeNeon ranks ahead of comparable models and #3 overall with a score of 61 on the Artificial Analysis Intelligence Index at its xhigh thinking level, landing just behind larger, slower models.
-
Google says Gemini 4 Neon ranks #1 overall on the Artificial Analysis Coding Agent Index, with a score of 58.6.
citeranking #1 overall on the Artificial Analysis Coding Agent Index (58.6)
-
Google says Gemini 4 Neon scores 65.2% on Terminal-Bench 4.1, and 46.2% on FrontierCode 2.0 at its default high setting.
Toned down to what the source says
citewith a score of 65.2% on Terminal-Bench 4.1 and 46.2% (at its default high setting) on FrontierCode 2.0.
-
Google says that on SWE-sweep Gemini 4 Neon leads on the mini-SWE harness with 12.8% and scores 39.0% in Google Antigravity, which Google describes as state-of-the-art.
citeOn SWE-sweep, it already leads on the mini-SWE harness (12.8%) and more than triples to a state-of-the-art 39.0% in Google Antigravity showing what the Antigravity harness unlocks.
-
Google says that at its default high thinking level Gemini 4 Neon pairs a high Artificial Analysis Intelligence Index score with a 32% lower cost per task than Carbon.
citeAt its default high thinking level, it pairs a high Artificial Analysis Intelligence Index score with a 32% lower cost per task than Carbon.
-
Google says that at xhigh thinking Gemini 4 Neon shows its highest intelligence score while costing 36%+ less per task than competing models at their highest setting.
citexhigh thinking adds deeper, multi-step reasoning for the hardest engineering, analytical, and scientific workloads. Neon shows its highest intelligence score here, while still costing 36%+ less per task than competing models at their highest setting.
-
Google says Gemini 4 Neon is rolling out in the Gemini API in Google AI Studio, in Gemini for business (via API and model dropdown), for paid Google AI Pro and Ultra subscribers in Google Antigravity, and for Pro and Ultra subscribers in the Gemini app.
citeGemini 4 Neon is rolling out today in the Gemini API in Google AI Studio, Gemini for business (via API and model dropdown), for paid Google AI Pro and Ultra subscribers in Google Antigravity, and for Pro and Ultra subscribers in the Gemini app.
-
Google says Gemini 4 Neon for cyber is rolling out to trusted defenders in the coming days as part of the Fairwind Program.
citeGemini 4 Neon for cyber is rolling out to trusted defenders in the coming days as part of the Fairwind Program.
-
Google prices Gemini 4 Neon at $3 per million input tokens ($0.15 per million cached input tokens, a 95% discount) and $15 per million output tokens, with a 1 million-token context window.
citeThe model is priced at $3 per million input tokens ($0.15 per million cached input tokens, a 95% discount) and $15 per million output tokens, with a 1 million-token context window.
2 Developer's guide to Gemini 4 Neon (Gemini Enterprise Agent Platform)
Open the source archived copy-
Google's developer guide says Gemini 4 Neon is its model built for complex coding, analytical reasoning and production-grade enterprise workflows.
citeGemini 4 Neon is our model built for complex coding, analytical reasoning, and production-grade enterprise workflows.
-
Google's developer guide says that while Carbon tackles the hardest long-horizon projects and Hydrogen handles fast everyday execution, Neon balances depth and efficiency.
citeWhile Carbon tackles your hardest, long-horizon projects and Hydrogen handles fast, everyday execution, Neon balances depth and efficiency and provides frontier-level reasoning for complex, high-frequency workflows quickly and efficiently.
-
Google's developer guide for Gemini 4 Neon shows a last-updated date of 2026-11-19 UTC.
citeLast updated 2026-11-19 UTC.
-
Gemini 4 Neon supports the thinking levels LOW, MEDIUM, HIGH (the default) and XHIGH, according to Google's developer guide.
citeThinking levels: Neon supports LOW, MEDIUM, HIGH (default), and XHIGH.
-
Google's developer guide says that, to maintain platform reliability, capacity and security, Gemini 4 Neon is available to verified enterprise customers in select regions.
citeNote: To maintain platform reliability, capacity, and security, Gemini 4 Neon is available to verified enterprise customers in select regions.
-
Google's developer guide gives Gemini 4 Neon a unified limit of 1,048,576 tokens that includes both input and output tokens, where output includes thinking, function calls and the final response.
citeUnified input and output token limit: Neon has a unified 1,048,576-token limit that includes both input and output tokens (where output includes thinking, function calls, and the final response).
-
Google's developer guide says every token the model processes is counted and billed, including thinking tokens.
citeToken counting and billing: Every token the model processes is counted and billed, including thinking tokens and internal tool system instructions and function declarations when using grounding.
Every sentence above is checked against these 2 sources. Each fact has its own address you can cite.
Source 1 · Google · 18 Nov 2026
Introducing Gemini 4 Neon
Google's launch post of 18 November 2026 says Gemini 4 Neon brings frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work and science workflows.
Introducing Gemini 4 Neon Nov 18, 2026 ... Today, we’re introducing Gemini 4 Neon — bringing frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work, and science workflows.
Google says Gemini 4 Neon ranks #3 overall with a score of 61 on the Artificial Analysis Intelligence Index at its xhigh thinking level, landing just behind larger, slower models.
Neon ranks ahead of comparable models and #3 overall with a score of 61 on the Artificial Analysis Intelligence Index at its xhigh thinking level, landing just behind larger, slower models.
Google says Gemini 4 Neon ranks #1 overall on the Artificial Analysis Coding Agent Index, with a score of 58.6.
ranking #1 overall on the Artificial Analysis Coding Agent Index (58.6)
Show all 10 factsShow fewer facts
Google says Gemini 4 Neon scores 65.2% on Terminal-Bench 4.1, and 46.2% on FrontierCode 2.0 at its default high setting.
Toned down to what the source sayswith a score of 65.2% on Terminal-Bench 4.1 and 46.2% (at its default high setting) on FrontierCode 2.0.
Google says that on SWE-sweep Gemini 4 Neon leads on the mini-SWE harness with 12.8% and scores 39.0% in Google Antigravity, which Google describes as state-of-the-art.
On SWE-sweep, it already leads on the mini-SWE harness (12.8%) and more than triples to a state-of-the-art 39.0% in Google Antigravity showing what the Antigravity harness unlocks.
Google says that at its default high thinking level Gemini 4 Neon pairs a high Artificial Analysis Intelligence Index score with a 32% lower cost per task than Carbon.
At its default high thinking level, it pairs a high Artificial Analysis Intelligence Index score with a 32% lower cost per task than Carbon.
Google says that at xhigh thinking Gemini 4 Neon shows its highest intelligence score while costing 36%+ less per task than competing models at their highest setting.
xhigh thinking adds deeper, multi-step reasoning for the hardest engineering, analytical, and scientific workloads. Neon shows its highest intelligence score here, while still costing 36%+ less per task than competing models at their highest setting.
Google says Gemini 4 Neon is rolling out in the Gemini API in Google AI Studio, in Gemini for business (via API and model dropdown), for paid Google AI Pro and Ultra subscribers in Google Antigravity, and for Pro and Ultra subscribers in the Gemini app.
Gemini 4 Neon is rolling out today in the Gemini API in Google AI Studio, Gemini for business (via API and model dropdown), for paid Google AI Pro and Ultra subscribers in Google Antigravity, and for Pro and Ultra subscribers in the Gemini app.
Google says Gemini 4 Neon for cyber is rolling out to trusted defenders in the coming days as part of the Fairwind Program.
Gemini 4 Neon for cyber is rolling out to trusted defenders in the coming days as part of the Fairwind Program.
Google prices Gemini 4 Neon at $3 per million input tokens ($0.15 per million cached input tokens, a 95% discount) and $15 per million output tokens, with a 1 million-token context window.
The model is priced at $3 per million input tokens ($0.15 per million cached input tokens, a 95% discount) and $15 per million output tokens, with a 1 million-token context window.
Source 2 · Google Cloud · 19 Nov 2026
Developer's guide to Gemini 4 Neon (Gemini Enterprise Agent Platform)
Google's developer guide says Gemini 4 Neon is its model built for complex coding, analytical reasoning and production-grade enterprise workflows.
Gemini 4 Neon is our model built for complex coding, analytical reasoning, and production-grade enterprise workflows.
Google's developer guide says that while Carbon tackles the hardest long-horizon projects and Hydrogen handles fast everyday execution, Neon balances depth and efficiency.
While Carbon tackles your hardest, long-horizon projects and Hydrogen handles fast, everyday execution, Neon balances depth and efficiency and provides frontier-level reasoning for complex, high-frequency workflows quickly and efficiently.
Google's developer guide for Gemini 4 Neon shows a last-updated date of 2026-11-19 UTC.
Last updated 2026-11-19 UTC.
Show all 7 factsShow fewer facts
Gemini 4 Neon supports the thinking levels LOW, MEDIUM, HIGH (the default) and XHIGH, according to Google's developer guide.
Thinking levels: Neon supports LOW, MEDIUM, HIGH (default), and XHIGH.
Google's developer guide says that, to maintain platform reliability, capacity and security, Gemini 4 Neon is available to verified enterprise customers in select regions.
Note: To maintain platform reliability, capacity, and security, Gemini 4 Neon is available to verified enterprise customers in select regions.
Google's developer guide gives Gemini 4 Neon a unified limit of 1,048,576 tokens that includes both input and output tokens, where output includes thinking, function calls and the final response.
Unified input and output token limit: Neon has a unified 1,048,576-token limit that includes both input and output tokens (where output includes thinking, function calls, and the final response).
Google's developer guide says every token the model processes is counted and billed, including thinking tokens.
Token counting and billing: Every token the model processes is counted and billed, including thinking tokens and internal tool system instructions and function declarations when using grounding.
Topics
The morning email
On the mornings we publish: the three top stories and up to four short ones. Free.
We email you a link to confirm. An issue may include one sponsor, always labelled Sponsored · Advertisement. Our emails count opens and clicks, not who made them. Unsubscribe in one click. What we keep
