Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. It is available on AI/ML API through the chat completions endpoint as google/gemini-4-argon.
Google reports Argon ahead of competing models from OpenAI and Anthropic on several benchmarks, including long-video understanding. These are vendor-reported results and have not been independently reproduced.
Sources: blog.google, venturebeat.com
Google describes Argon as built for agentic software engineering, including analysing fleet-wide profiling telemetry and migrating C/C++ codebases with agents running over long horizons.
It is also positioned for deep mathematical and scientific problem solving, and for autonomous multi-step agents that keep working across many sequential tool calls.
The 1,048,576-token context window suits long-document and multi-file analysis, and image input allows multimodal document and screenshot understanding.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "google/gemini-4-argon", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "google/gemini-4-argon", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"google/gemini-4-argon","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 52.6 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 30, 2026 |
| DeepSWE | 77.9% | Agentic software engineering: resolving real repository tasks end to end | Source | October 7, 2026 |
| Terminal-Bench 4 | 57.4% | Autonomous shell and terminal task completion, fourth release | Source | October 7, 2026 |
| AutomationBench | 51.3% | Completion of routine automation workflows | Source | October 7, 2026 |
Figures reported by Google; not independently reproduced.
Yes. Gemini 4 Argon is live on AI/ML API and callable with the model id google/gemini-4-argon on the chat completions endpoint, using the same API key as every other model on the platform.
Up to 65,536 tokens per response, within a 1,048,576-token context window.
On AI/ML API, Gemini 4 Argon is priced at $2.60 per 1M input tokens and $13.00 per 1M output tokens, with cached input at $0.13 per 1M tokens.
Gemini 4 Argon has a 1,048,576-token context window and can return up to 65,536 tokens in one response.
Google positions Argon for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. It cites agents analysing fleet-wide profiling telemetry and migrating C/C++ codebases over long horizons. The 1M-token context window also suits long-document and multi-file analysis.
Create an AI/ML API account, generate a key in Key Management and call google/gemini-4-argon on https://api.aimlapi.com/v1/chat/completions. No separate Google access is needed.
No. Google reports Argon ahead of competing models from OpenAI and Anthropic on several benchmarks, including long-video understanding. Those results are vendor-reported and have not been independently reproduced.
