gemini-4-argonGemini 4 Argon is Google's flagship reasoning model — announced on 30 September 2026 and open to developers on the paid tier of the Gemini API from 13 October 2026 — and on APIMODELS it costs $1.10 per 1M input tokens and $5.50 per 1M output tokens, 45% below Google's $2 / $10 introductory list, with cached input at $0.055 (Google: $0.10). Google built Argon for long-horizon work that needs deep reasoning and structural precision: real-world software engineering and code migrations, enterprise knowledge work such as legal and finance, frontier research and cybersecurity defense. The Gemini 4 Argon benchmarks Google published: 77.9% on DeepSWE v1.1, which Google calls a new state of the art for real-world software engineering; 91.7% on LVBench long-video understanding; 68% on CWE-bench v1 vulnerability remediation, tied for first; and 51.3% on AutomationBench, ranked first, for end-to-end business tasks. These are Google's own figures, so run your own prompts before moving a production workload. Specifications: a 1M-token context window with output allowed to run up to 1M tokens in one response (earlier Gemini models stopped at 64K), text, image and video input, and tool calling. Thinking tokens bill at the output rate, and Argon thinks even on trivial prompts — in our own test, replying "OK" used 66 thinking tokens — so budget by total output rather than visible text. Caching is implicit only: repeated prompt prefixes are discounted automatically and billed at $0.055 per 1M, but explicit cache objects are not supported. On pricing: Google's $2 / $10 is an introductory rate that runs through 30 November 2026, and from 1 December 2026 Google lists Argon at $4 / $20 (cached $0.20); the price shown on this page is always the one that bills. Call it with model gemini-4-argon on the native Gemini endpoint or the OpenAI-compatible /v1/chat/completions, one API key for both; requests switch to another upstream channel automatically if one fails, and failed requests are not charged. When not to use it: for short, high-volume calls Gemini 3.8 Flash ($0.450 / $2.250) is the more affordable Google model here.
View complete API reference with all parameters and examples.
View complete API reference with streaming, thinking, and more.
Billing: Cost = (input_tokens * input_price + output_tokens * output_price) / 1,000,000
$1.10 / $5.50 per 1M tokens against Google's $2 / $10 introductory list; cached input $0.055
Google: DeepSWE v1.1 77.9%, LVBench 91.7%, CWE-bench v1 68% (tied first), AutomationBench 51.3% (first)
Google: long-horizon coding and code migrations, legal and finance work, research, cybersecurity defense
Output can run to 1M tokens in one response (64K on earlier Gemini models); text, image and video input; tool calling
Our test: replying "OK" used 66 thinking tokens; caching is implicit only, cached reads $0.055 per 1M
Google moves Argon from $2 / $10 to $4 / $20 per 1M tokens on 1 December 2026; this page always shows what bills
Gemini 4 Argon is a Large Language Model API provided by Google. Gemini 4 Argon is Google's flagship reasoning model — announced on 30 September 2026 and open to developers on the paid tier of the Gemini API from 13 October 2026 — and on APIMODELS it costs $1.10 per 1M input tokens and $5.50 per 1M output tokens, 45% below Google's $2 / $10 introductory list, with cached input at $0.055 (Google: $0.10). Google built Argon for long-horizon work that needs deep reasoning and structural precision: real-world software engineering and code migrations, enterprise knowledge work such as legal and finance, frontier research and cybersecurity defense. The Gemini 4 Argon benchmarks Google published: 77.9% on DeepSWE v1.1, which Google calls a new state of the art for real-world software engineering; 91.7% on LVBench long-video understanding; 68% on CWE-bench v1 vulnerability remediation, tied for first; and 51.3% on AutomationBench, ranked first, for end-to-end business tasks. These are Google's own figures, so run your own prompts before moving a production workload. Specifications: a 1M-token context window with output allowed to run up to 1M tokens in one response (earlier Gemini models stopped at 64K), text, image and video input, and tool calling. Thinking tokens bill at the output rate, and Argon thinks even on trivial prompts — in our own test, replying "OK" used 66 thinking tokens — so budget by total output rather than visible text. Caching is implicit only: repeated prompt prefixes are discounted automatically and billed at $0.055 per 1M, but explicit cache objects are not supported. On pricing: Google's $2 / $10 is an introductory rate that runs through 30 November 2026, and from 1 December 2026 Google lists Argon at $4 / $20 (cached $0.20); the price shown on this page is always the one that bills. Call it with model gemini-4-argon on the native Gemini endpoint or the OpenAI-compatible /v1/chat/completions, one API key for both; requests switch to another upstream channel automatically if one fails, and failed requests are not charged. When not to use it: for short, high-volume calls Gemini 3.8 Flash ($0.450 / $2.250) is the more affordable Google model here. Through APIMODELS platform, you can access this model via a unified API with transparent pay-as-you-go pricing. Current pricing: Input: $1.10, Output: $5.50 per 1M tokens.
Build intelligent conversational systems to automatically answer user queries and improve service efficiency.
Automatically write articles, emails, ad copy, and other text content to boost productivity.
Assist with code writing, debugging, and code review to accelerate software development.
Understand and analyze unstructured data, extract key insights, and generate summary reports.
Gemini 4 Argon is available through APIMODELS at: Input: $1.10, Output: $5.50 per 1M tokens. Billing is pay-as-you-go — you only pay for what you generate.
Sign up at APIMODELS, get your API key, and call our unified API endpoint. We provide detailed API documentation with code examples in cURL, Python, and Node.js.
APIMODELS offers the same Gemini 4 Argon model through our aggregation platform. We provide a unified API interface so you do not need separate accounts for each provider - one API key to access all models.
$1.10 per 1M input tokens and $5.50 per 1M output, with cached input at $0.055 — 45% below Google's introductory list of $2 / $10 and $0.10 cached. Thinking tokens bill at the output rate, as they do at Google, and there is no long-context surcharge: Google lists one rate for the whole 1M window. Google's introductory rate runs through 30 November 2026; from 1 December 2026 Google lists Argon at $4 / $20, with cached input at $0.20. The price shown on this page is always the one that bills. You pay only for the tokens used, failed requests are not charged, and there is no subscription.
Google's launch figures: 77.9% on DeepSWE v1.1, which Google calls a new state of the art for real-world software engineering; 91.7% on LVBench, a long-video understanding benchmark; 68% on CWE-bench v1 for vulnerability remediation, tied for first; and 51.3% on AutomationBench, which measures end-to-end business tasks, ranked first. All four are Google's own numbers rather than independent runs, so treat them as a starting point and compare on your own prompts — switching models here is a change to the model field.
Pick Argon for long, hard tasks where a correct answer matters more than latency — multi-step coding and code migrations, long-video analysis, research, security work — and for anything that needs a very long single output, since it can write up to 1M tokens in one response. On APIMODELS it costs $1.10 / $5.50 per 1M tokens. Gemini 3.1 Pro Preview costs $0.882 / $5.294 but is a preview build of the previous generation; Gemini 3.8 Flash at $0.450 / $2.250 is the better fit for high-volume agent loops and short requests. All three share one API key and the same endpoints.
Set the model to gemini-4-argon and authenticate with your apimodels key. In the native Gemini format, POST https://api.apimodels.app/v1beta/models/gemini-4-argon:generateContent (streamGenerateContent for streaming) with a contents array. In the OpenAI format, POST https://api.apimodels.app/v1/chat/completions, so the OpenAI SDK works once base_url is https://api.apimodels.app/v1. Streaming, multimodal input and function calling pass through. You do not need a Google Cloud project or billing account, it works from mainland China, and if one upstream channel fails the request switches to another automatically.
Because thinking tokens bill at the output rate, and Argon thinks before it answers even on trivial prompts: in our own test, asking it to reply "OK" used 66 thinking tokens. Budget by candidatesTokenCount plus thoughtsTokenCount in usageMetadata, not by the length of the text you see, and set maxOutputTokens (max_tokens in the OpenAI format) on open-ended tasks. For long prompts you send repeatedly, caching helps: Argon supports implicit caching only, so repeated prefixes are discounted automatically and billed at $0.055 per 1M, but explicit cache objects cannot be created.
A 1M-token context window, with output allowed to run up to 1M tokens in a single response — earlier Gemini models stopped at 64K — plus text, image and video input and tool calling. At Google it is available only on the paid tier of the Gemini API, with no free tier; here it is pay-as-you-go from the same balance as every other model. Long outputs bill at $5.50 per 1M tokens like any other output, so cap maxOutputTokens when you do not need the full length.
On APIMODELS, Gemini 4 Argon runs alongside 60+ models on one API key and one balance, so choosing is about fit, not lock-in. It supports 45% Below Official, Flagship Reasoning, 1M Context, Up to 1M Output, and you can weigh it on price and capability against other Large Language Model models, then switch by changing a single model-name string — no new account or integration. Browse every Large Language Model option with live pricing at apimodels.app/models.
Gemini 4 Argon supports: 45% Below Official, Flagship Reasoning, 1M Context, Up to 1M Output. See the APIMODELS docs for full parameters and call examples.
Yes. APIMODELS exposes Gemini 4 Argon through a single unified API and one key — no separate provider accounts, and no need to handle each provider's regional network access yourself.
We support Stripe (Visa, Mastercard, and other international cards) and Alipay. Credits are available instantly after payment.