Gemini 4 Neon’s Value Equation: What The Table Shows And Leaves Unsaid
AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: Gemini 4 Neon’s Value Equation: What The Table Shows And Leaves Unsaid on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get smart everyday buys delivered free with Prime

  • Fast, free delivery on millions of items
  • Prime Video, Amazon Music and more included
  • Member-only deals all year
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

TL;DR

Google released Gemini 4 Neon on November 18, describing it as a model balancing depth and efficiency. Artificial Analysis measured a one-point intelligence-score gain over Gemini 4 Carbon at Neon’s highest setting, alongside lower cost and shorter task times; the results vary by effort level and benchmark. Google’s own table compares Neon with Claude Opus 5.5 and GPT-6.2 Sol, but not with two models that rank ahead of Neon on the independent index.

Google released Gemini 4 Neon on November 18, presenting it as a model that balances depth and efficiency. In data from independent evaluator Artificial Analysis, Neon scored 61 on the Intelligence Index at its highest effort setting, one point above Gemini 4 Carbon at high; the same comparison showed lower cost per index task and shorter completion time, but did not establish that Neon leads the strongest competing models.

Artificial Analysis reports that a Neon xhigh index task costs $6.44 and takes 21.5 minutes, compared with $7.76 and 37.1 minutes for Carbon at high, using standard list prices. At Neon’s default high setting, its score is 60—equal to Carbon’s high score—while a task costs $3.74 and takes 14.8 minutes. The evaluator described the release as “a small Intelligence gain over Gemini 4 Carbon at a ~17% lower Cost per Task.”

The results change substantially with Neon’s effort setting. Artificial Analysis lists scores of 61 at xhigh, 60 at high, 55 at medium and 44 at low; the corresponding cost per index task falls from $6.44 at xhigh to $0.75 at low. Going from high to xhigh adds one index point but raises the reported task cost from $3.74 to $6.44. These figures describe the evaluator’s index workload, not a guaranteed price or completion time for every use.

Google lists Neon at $3 per million input tokens and $15 per million output tokens, below Carbon’s standard $4 and $20 rates. However, Carbon’s listed rates are discounted to $2 and $10 through November 30, according to the source article. Artificial Analysis attributes the task-cost difference to lower token prices and output generation of about 110 tokens per second for Neon versus 55 for Carbon, partly offset by Neon using about 15% more output tokens per task.

Performance is not uniformly better across tests. Neon xhigh scored 63.6% on Terminal-Bench 4.0, below Carbon high at 65.2%. On Artificial Analysis’s Coding Agent Index, Neon scored 58.6, close to Claude Fable 5.2 at 58.5 and Claude Opus 5.5 at 58.2. The reported task cost was $6.86 for Neon, compared with $15.47 for Fable and $12.61 for Opus.

At a glance
reportWhen: Released November 18, 2026; benchmark f…
The developmentGoogle released Gemini 4 Neon, and independent benchmark data show a modest score gain over Carbon alongside lower per-task costs and faster completion times.

Gemini 4 Neon

Four models in seven weeks

Google’s Gemini 4 releases side by side, each at its best tested setting in the independent Artificial Analysis index. The best value in each row is outlined.

ArgonCarbonHydro.Neon ReleasedSep 30Oct 28Nov 12Nov 18 Index50605461 Cost/task$4.97$7.76$3.16$6.44 Min./task29.737.115.321.5 Output $$10$20$7.50$15 Carbon $2/$10 discount ends Nov 30Neon: +1 point, $1.32 less per task
  1. Four models in seven weeksArgon on September 30, Carbon on October 28, Hydrogen on November 12, Neon on November 18. Neon is adapted from Carbon.
  2. A small step in the indexNeon at xhigh scores 61, one point above Carbon at high. It ranks third among model families, behind GPT-6.2 Astra and Claude Fable 5.2.
  3. A bigger step in cost and timeOne index task costs $6.44 instead of $7.76 and takes 21.5 instead of 37.1 minutes. Neon’s tokens are 25% cheaper than Carbon’s standard price.
  4. What to watchUntil November 30, Carbon still sells at a discounted $2/$10. Google’s own table compares Neon with Claude Opus 5.5 and GPT-6.2 Sol, not with the index leaders.

Source: Artificial Analysis, Gemini 4 Neon model page and article, accessed Nov 19, 2026 (Intelligence Index v5.0.1, cost and time per task at standard prices, release dates); Google, “Introducing Gemini 4 Neon”, Nov 18, 2026; Gemini 4 Neon model card. Output price in US$ per million tokens, standard rates. Redrawn by Thorsten Meyer AI.

Graphic: Powered by Thorsten Meyer AI

Cost Gains Depend on the Workload

For teams running repeated, multi-step model tasks, price and elapsed time per task can matter as much as a headline benchmark score. The Artificial Analysis figures suggest Neon may reduce both relative to Carbon in the evaluator’s index workload, giving developers a reason to test it where throughput and operating costs matter. The source figures are historical observations at the prices and settings specified, not a promise of future costs or performance.

The trade-off is that a model choice cannot be settled by one aggregate score. Neon’s effort level changes its measured score and cost, and results differ across tasks: it is close to leading coding models on one index but trails Carbon on Terminal-Bench 4.0. Buyers evaluating it should compare the settings and benchmarks that resemble their own use, and account for any temporary discounts when estimating costs.

Google’s published comparison also has a scope limitation. It reports Neon ahead in 16 of 19 rows against Claude Opus 5.5 and GPT-6.2 Sol, with those rivals ahead in three rows. But the source article says the table omits Claude Fable 5.2 and GPT-6.2 Astra, which rank above Neon on Artificial Analysis’s Intelligence Index. That makes the table useful for understanding Google’s chosen comparisons, but not a complete view of the model field.

Vitamix 5200 Professional-Grade Blender with 64 oz Container, Black

Vitamix 5200 Professional-Grade Blender with 64 oz Container, Black

Apple AirPods Pro 3 Wireless Earbuds with Active Noise Cancellation

Apple AirPods Pro 3 Wireless Earbuds with Active Noise Cancellation

As an affiliate, we earn on qualifying purchases.

Neon Builds on Gemini 4 Carbon

Neon is the fourth Google model release in about seven weeks, following Gemini 4 Argon on September 30, Carbon on October 28 and Hydrogen on November 12. Google’s launch post describes Carbon as suited to harder, long-horizon projects and Hydrogen to fast everyday execution, placing Neon between them. Its model card says Neon is adapted from Carbon and refers readers to Carbon’s documentation for architecture, training data and hardware details.

Google says Neon accepts text, images, audio and video, produces text, and has a one-million-token context window. Its default output limit is 64,000 tokens, with up to one million through background long decode. The listed knowledge cutoff is March 2026, and users can choose among four effort levels, with high as the default. Google lists availability through the Gemini API in AI Studio and several Gemini products for eligible subscribers; a cyber-defense version is being made available to eligible members of its Fairwind Program.

Google’s launch post calls Neon “frontier-level intelligence with fast end-to-end task completion.” It also says Neon had its “lowest fabrication rate yet in internal Google Antigravity testing.” These are Google’s descriptions of the model and internal results; the fabrication-rate claim uses a different measurement from the independent evaluator’s hallucination test.

““frontier-level intelligence with fast end-to-end task completion.””

— Google, in its Gemini 4 Neon launch post

Limits of the Published Comparisons

The available figures do not show how Neon performs across every real-world workload, or whether the evaluator’s task costs and completion times will hold for different prompts, usage patterns or settings. The source material does not provide a complete independent comparison across all current leading models. In particular, Google’s 19-row table excludes Claude Fable 5.2 and GPT-6.2 Astra, while the independent index places those models ahead of Neon.

There is also no direct basis in the material for equating Google’s internal fabrication-rate result with Artificial Analysis’s hallucination measure. In the latter, Neon’s reported hallucination rate on AA-Omniscience was 14.4%, versus Carbon’s 11.4%; the source says several other models had higher rates in that test. Different tests may measure different behaviors, so these values do not verify or refute Google’s internal claim.

The source article does not supply full details for every benchmark in Google’s table or an independent audit of Google’s internal testing. It also does not say how long Neon’s listed prices will remain unchanged. The Carbon discount has a stated end date of November 30, but the material does not establish what pricing will apply after that date.

Testing Neon Against Real Tasks

The next useful evidence will come from broader independent comparisons and from developers testing Neon on their own workloads at the effort levels they intend to use. Those evaluations can show whether the reported cost and speed advantages persist when task type, output length and model setting change, and whether trade-offs on particular benchmarks matter in practice.

Readers comparing prices should also check current rates and the status of Carbon’s temporary discount after November 30. Google says eligible Fairwind Program members will receive access to the cyber-defense version “over the next few days.” The source material does not give a further release date or announce a next benchmark milestone.

Key Questions

When did Google release Gemini 4 Neon?

Google released Gemini 4 Neon on November 18, 2026, according to the source material.

Does Neon score higher than Gemini 4 Carbon?

At its highest listed setting, Neon scored 61 on Artificial Analysis’s Intelligence Index, one point above Carbon at high. At Neon’s default high setting, both scored 60. The comparison depends on effort setting and benchmark.

Is Gemini 4 Neon cheaper to run?

Artificial Analysis reported lower cost per index task for Neon than for Carbon at standard list prices: $6.44 at Neon xhigh versus $7.76 for Carbon high. Neon high was listed at $3.74 per task. Actual costs depend on usage, settings and current pricing.

Does Google’s benchmark table compare Neon with the top-ranked models?

Not all of them. The source says Google’s 19-row table compares Neon with Claude Opus 5.5 and GPT-6.2 Sol, but omits Claude Fable 5.2 and GPT-6.2 Astra, which rank ahead of Neon on the Artificial Analysis Intelligence Index.

Where can users access Gemini 4 Neon?

Google lists access through the Gemini API in Google AI Studio and eligible Gemini products and subscriptions. The source also says a cyber-defense version is being offered to eligible Fairwind Program members.

Source: ThorstenMeyerAI.com

This content is for general information only and is not financial, tax or legal advice. Consult a qualified professional for decisions about your money.
You May Also Like

Adviser Fees Equity Release: The Costs That Add Up Faster Than People Expect

When it comes to equity release adviser fees, understanding hidden costs can save you money—discover how to avoid surprises and manage expenses effectively.

Equity Release vs Retirement Interest Only Mortgage: The Difference That Could Save You a Costly Mistake

The key differences between equity release and retirement interest-only mortgages could save you from costly mistakes—discover which option is right for your financial future.

Equity Release vs Reverse Mortgage: The Difference That Could Save You a Costly Mistake

Navigating equity release and reverse mortgages can be confusing—understanding their differences could save you from costly mistakes; discover which option is best for you.