Gemini 4 Neon

Released
2026-11-18
Last refreshed
2026-11-18
Status
Researched 1d ago
ProprietaryCommercial use: restrictedMultimodalCodingRAGAgentsLong contextVisionJSON / Tool useCodingAgentMultimodal
Google DeepMind releases · 54 in the last 12 months · this family litChangelog →
Specifications
Family
Gemini 4
Released
2026-11-18
Context
1M
Max output
1M
Knowledge cutoff
2026-03
Specialization
general
Openness
Proprietary
License
ProprietaryRestricted
Weights
Not released
Code
Unknown
Created by

Pioneering artificial intelligence research.

London, United Kingdom
Founded 2014
Website
Pricing
Output / 1M
$15.00
Input / 1M
$3.00

Cheapest of 4 routes · Google AI Studio · cache read $0.150

About

Gemini 4 Neon is Google DeepMind's Gemini 4 model, released generally available on November 18, 2026 in the Gemini API, Google AI Studio, and Vertex AI (Gemini Enterprise Agent Platform). Google positions it for complex software engineering, knowledge work, and science workflows. The model card says it is adapted from Gemini 4 Carbon. It takes text, image, video, audio and PDF input and returns text, with a unified 1,048,576-token context window and long decode up to 1,048,576 output tokens (65,536 by default).

Providers(4)

Compare all 4

Compare API pricing across 4 tracked provider routes for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MBatch in / outCacheRoute
Google AI Studio$3.00$15.00$1.50 / $7.50read $0.150
Serverless
GCP Vertex AI$3.00$15.00$1.50 / $7.50read $0.150
Serverless
OpenRouter$3.00$15.00-read $0.150
Serverless
Vercel AI Gateway$3.00$15.00-read $0.150
Serverless

Available via routers & gateways(13)

LiteLLM

Gateway

Open-source Python SDK and proxy server that unifies 100+ LLM APIs behind a single OpenAI-compatible interface, with load balancing, cost tracking, and configurable failover.

Free OSSGoogle AI StudioGCP Vertex AI

OpenRouter

Hybrid

Unified hybrid gateway to 500+ models from 80+ providers via a single OpenAI-compatible API, with an optional Auto Router that selects a model per prompt from market spend data.

Passthrough + feeGoogle AI StudioGCP Vertex AI

Portkey

Gateway

Production AI gateway routing to 1,600+ LLMs with failover, load balancing, caching, and guardrails; MIT-licensed open-source core is self-hostable. Now part of Palo Alto Networks as Prisma AIRS AI Gateway.

SubscriptionGoogle AI StudioGCP Vertex AI

AIRouter

Router

Commercial LLM router (branded "AI Router" on its site) that picks the best model per request via a Decision API or an OpenAI-compatible Chat Completions endpoint, with a private selection mode that routes on anonymized patterns; sold as BYOK subscription plans priced by routed tokens.

SubscriptionGoogle AI StudioGCP Vertex AI

Helicone

Gateway

Observability-first AI gateway with routing, caching, rate limiting, and request tracing; Apache-2.0 open-source platform with a managed hosted tier for logging and analytics.

SubscriptionGoogle AI StudioGCP Vertex AI

Kong AI Gateway

Gateway

Multi-LLM AI gateway built on Kong Gateway 3.x, adding semantic routing, load balancing, guardrails, and MCP traffic analytics as plugins over Kong's existing API management platform.

SubscriptionGoogle AI StudioGCP Vertex AI

Capabilities

VisionMultimodalReasoningJSON / Tool useStructured OutputsCode ExecutionPrompt CachingBatch APIAudioVideo

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Benchmark scores(6)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
46.2FrontierCode 2.0 Main scoreObserved 2026-11-18
Configuration: gemini-4-neon (high)
Harness: Official public leaderboard; Google's evals methodology says Gemini 4 Neon results use high thinking
Evaluator: Cognition
Confidence: self-reported
Notes: Google DeepMind model card results table (as of November 2026): FrontierCode 2.0 Main score 46.2%. Evals methodology: sourced from the official public leaderboard; Neon uses high thinking. The Cognition board was not independently recaptured in this packet. Timestamp is retrieval date.
56.3AutomationBenchObserved 2026-11-18
Configuration: Gemini 4 Neon
Harness: Official public leaderboard (Zapier), as stated in Google's evals methodology
Evaluator: Zapier
Confidence: self-reported
Notes: Google DeepMind model card results table (as of November 2026): AutomationBench 56.3%. Evals methodology: sourced from the official public leaderboard. The owner board was not independently recaptured in this packet. Timestamp is retrieval date.
75.8Agents' Last Exam v1.1Observed 2026-11-18
Configuration: Agents' Last Exam v1.1; Antigravity-CLI 1.4.8; 122-task; 5-hour window; safety filters enabled; averaged rubric score
Harness: Self-computed on the 122-task v1.1 benchmark using the Antigravity-CLI version 1.4.8 harness through official integration and a 5-hour window, with safety filters enabled; averaged rubric score
Evaluator: Google DeepMind
Confidence: self-reported
Notes: Google DeepMind model card results table (as of November 2026): Agents' Last Exam v1.1 75.8%. Evals methodology records the self-computed harness above. Timestamp is retrieval date.
70.5Terminal-Bench-Science 0.1Observed 2026-11-18
Configuration: Gemini 4 Neon; mini-swe-agent vision; average of 5 attempts; Average Reward
Harness: Self-computed using a mini-swe-agent vision harness, averaged over 5 attempts per task (Average Reward)
Evaluator: Google DeepMind
Confidence: self-reported
Notes: Google DeepMind model card results table (as of November 2026): Terminal-Bench-Science 0.1 70.5%. Evals methodology matches this harness. Timestamp is retrieval date.
90.5SEC-bench ProObserved 2026-11-18
Configuration: Pass@1; 183-task May 2026 JavaScript engine split; average of 3 independent attempts; Antigravity harness
Harness: Self-computed using the open source evaluation setup; Pass@1 on the 183-task May 2026 JavaScript engine split, averaged over 3 independent attempts per task, using the Antigravity harness
Evaluator: Google DeepMind
Confidence: self-reported
Notes: Google DeepMind model card results table (as of November 2026): SEC-bench Pro 90.5%. Evals methodology records the self-computed protocol above. Timestamp is retrieval date.
73.5ChartographyObserved 2026-11-18
Configuration: Gemini 4 Neon; self-computed without tools
Harness: Self-computed without tools following the official Surge public leaderboard
Evaluator: Google DeepMind
Confidence: self-reported
Notes: Google DeepMind model card results table (as of November 2026): Chartography 73.5%. Evals methodology: Neon results are self-computed without tools. Timestamp is retrieval date.

Migration checks

No linked migration route is available for this model yet.

API versions

gemini-4-neongoogle/gemini-4-neon