{"schema":"postcutoff/model@1","as_of":"2026-11-15T00:33:00+01:00","url":"https://postcutoff.com/m/gemini-4-argon/","md":"https://postcutoff.com/m/gemini-4-argon/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"gemini-4-argon","name":"Gemini 4 Argon","org":"Google DeepMind","family":"Gemini 4","released":"2026-10-13","status":"current","type":"reasoning-llm","modality_in":["text","image","audio","video","pdf"],"modality_out":["text"],"open_weights":false,"model_license":"proprietary","context_window":1048576,"max_output":65536,"knowledge_cutoff":"2026-03","pricing":{"input":2,"output":10,"cached_input":0.1,"unit":"per 1M tokens, paid tier only, introductory price through 2026-11-30; from 2026-12-01: $4.00 input / $20.00 output / $0.20 cached input. Output price includes thinking tokens. Batch and Flex: half price. Priority: $3.60 / $18.00 (then $7.20 / $36.00)","source":"https://ai.google.dev/gemini-api/docs/pricing"},"price_line":"$2 in, $10 out per 1M tokens","access":[{"provider":"Gemini API, generateContent (paid tier only; works with google_search grounding and audio input, tested 2026-10-25; the Interactions API also works)","model_id":"gemini-4-argon","endpoint":"https://generativelanguage.googleapis.com/v1beta/models/gemini-4-argon:generateContent","docs":"https://ai.google.dev/gemini-api/docs/google-search"},{"provider":"Gemini API, Interactions API (paid tier only; no Free tier; the form Google's docs samples use)","model_id":"gemini-4-argon","endpoint":"https://generativelanguage.googleapis.com/v1beta/interactions","docs":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon"},{"provider":"Gemini Enterprise Agent Platform (Google Cloud, location global)","model_id":"gemini-4-argon","endpoint":"https://aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/global/publishers/google/models/gemini-4-argon:generateContent","docs":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/guides/gemini-4-argon"},{"provider":"Google AI Studio (paid API tier or Google AI Ultra)","url":"https://aistudio.google.com/prompts/new_chat?model=gemini-4-argon"},{"provider":"Web app (Gemini app; Google AI Pro and Ultra subscribers)","url":"https://gemini.google.com"},{"provider":"AI Mode in Google Search (Google AI Pro and Ultra subscribers)","url":"http://aimode.google/"},{"provider":"Google Antigravity (Google AI Ultra subscribers and eligible enterprise plans)","url":"https://antigravity.google/"},{"provider":"GitHub Copilot (Pro+, Max, Business, Enterprise; generally available since 2026-10-13, gradual rollout, billed at provider list pricing)","url":"https://github.blog/changelog/2026-10-13-gemini-4-argon-in-github-copilot/"},{"provider":"OpenRouter ($2 / $10 per 1M tokens; also google/gemini-4-argon:batch at $1 / $5; 1,048,576-token context; checked 2026-10-14)","model_id":"google/gemini-4-argon","url":"https://openrouter.ai/google/gemini-4-argon"},{"provider":"Snowflake Cortex AI (public preview since 2026-10-13; Cortex AI Functions, cross-region inference; CoWork, CoCo and Cortex Agents \"coming soon\"; model id not stated)","url":"https://www.snowflake.com/en/blog/gemini-4-argon-snowflake-cortex-ai/"},{"provider":"Google DeepMind Fairwind Program (vetted cyber defenders; no cyber guardrails)","model_id":"gemini-4-argon-cyber-permissive","url":"https://deepmind.google/fairwind-program/"}],"capabilities":[{"name":"Unified 1M-token window with long decode","detail":"Input, thinking and output share one 1,048,576-token window. One call returns up to 65,536 tokens (32,768 for non-streaming direct HTTP on the enterprise platform); long decode continues in 64K hops up to 1M tokens minus the input, via background execution in the Interactions API or a continuation_token in generateContent. Earlier Gemini models capped output at 64K.","first":true,"discovered":"launch","source":"https://ai.google.dev/gemini-api/docs/long-decode"},{"name":"Enterprise knowledge work","detail":"Vendor-reported: Vals Index 68.9% (#1; confirmed on the Vals AI leaderboard as of Oct 12, 2026), Harvey Legal Agent Benchmark 19.6% vs 6.7% for the next model, Vals Finance Agent v2 65.4%, Zapier AutomationBench 51.3% (#1).","first":false,"discovered":"launch","source":"https://deepmind.google/models/model-cards/gemini-4-argon/"},{"name":"Low hallucination rate","detail":"Artificial Analysis measured a 15% hallucination rate on AA-Omniscience, the lowest of any model scoring 45+ on its Intelligence Index at the time (GPT-6 Astra 51%). Measured on the Sept 30 pre-release version.","first":false,"discovered":"launch","source":"https://x.com/ArtificialAnlys/status/2105392625788637299"},{"name":"Frontier software engineering (mixed)","detail":"Vendor-reported DeepSWE v1.1 77.9% (Opus 5.5 74.2%, GPT-6 Astra 74.1%), but it trails on FrontierSWE v2 (55.0% vs Astra 65.5%) and Terminal-Bench 4.0 (57.4% vs Opus 5.5 66.4%). Independent: Cognition FrontierCode 1.1 60.5%, level with GPT-6.1 Sol; Every found it weak at refactoring existing code.","first":false,"discovered":"launch","source":"https://devin.ai/blog/gemini-4-argon"},{"name":"Autonomous vulnerability discovery and patching","detail":"CWE-bench v1 68% pass@1, tied first with GPT-6 Astra and Grok 4.7 (public leaderboard). A separate id, gemini-4-argon-cyber-permissive, serves Fairwind-approved defenders without cyber guardrails. Google's model card says it cannot rule out that Argon reached its cybersecurity critical capability level.","first":false,"discovered":"launch","source":"https://deepmind.google/models/model-cards/gemini-4-argon/"},{"name":"Long-context and long-video understanding","detail":"Vendor-reported GraphWalks BFS 256K–1M 84.2% (Astra 71.8%) and LVBench 91.7%. Agentic video understanding is generally available with this model (off by default).","first":false,"discovered":"launch","source":"https://ai.google.dev/gemini-api/docs/latest-model"},{"name":"Embodied and spatial reasoning","detail":"Vendor-reported 87% on the ERQA embodied-reasoning benchmark, which Google says is the best of any frontier model; used in a video-to-simulation-to-robot pipeline on Apptronik's Apollo humanoid.","first":false,"discovered":"launch","source":"https://developers.googleblog.com/gemini-4-argon-for-robotics/"}],"entry":"2026-10-13-gemini-4-argon-general-availability","notes":"Successor: Gemini 4 Carbon (announced 2026-10-28, Fairwind cyber defenders only at first; see gemini-4-carbon), #1 on Artificial Analysis Intelligence Index v5 at 59, where Argon (high) scores 50; Argon stays the callable Gemini 4 model until Carbon is generally available. Generally available since 2026-10-13 (announced 2026-09-30 for Fairwind cyber defenders only; see 2026-09-30-gemini-4-argon). Knowledge cutoff per Google's model card: 'The knowledge cutoff for Gemini 4 is March 2026, but in some domains the model's knowledge may be less up to date.' max_output is the long-decode ceiling; a single synchronous call returns at most 65,536 tokens, and the API's models.get metadata reports outputTokenLimit 32768 (checked 2026-10-13). Thinking levels low / medium (default) / high; temperature, top_p and top_k are ignored; explicit context caching, the Live API and image or audio output are not supported. Not on the API Free tier; API and AI Studio access is not included in the Google AI Pro plan, though Pro subscribers get the model in the Gemini app and AI Mode. Frontier Safety Framework: Google cannot rule out the biological CBRN CCL, the cybersecurity CCL and the stealth/situational-awareness TCL, and says safeguards lower the risk below each. Independent: Artificial Analysis Intelligence Index 53 (Sept 30 version), Arena Text #1 at 1525 on Sept 30 (not re-checked). Safety: Google's 217-page Safety & Security Report (Oct 13, 2026, https://storage.googleapis.com/deepmind-media/gemini/gemini_4_safety_report.pdf) says it cannot rule out the unmitigated model reaching its cyber and biological critical capability levels and its stealth and situational-awareness level; external testers included UK AISI, the US NIST centre, Faculty and Apollo Research. Ids and prices re-checked against ai.google.dev on the evening of 2026-10-13: unchanged. generateContent checked 2026-10-25: models.get lists generateContent, countTokens and batchGenerateContent, and one live v1beta generateContent call with a WAV audio part plus the google_search tool answered the spoken question and returned groundingMetadata (web search queries and 2 chunks); Google's docs samples use the Interactions API, and the Google Search grounding page lists Gemini 4 Argon as supported.","verified":"2026-10-25","body_md":"Google's first Gemini 4 model and its flagship reasoning model. Model id `gemini-4-argon` (stable). Google says more Gemini 4 family\nmodels will follow, without names or dates.\n\nQuick start (Interactions API, from Google's docs):\n\n```bash\ncurl \"https://generativelanguage.googleapis.com/v1beta/interactions\" \\\n  -H \"x-goog-api-key: $GEMINI_API_KEY\" -H 'Content-Type: application/json' -X POST \\\n  -d '{\"model\": \"gemini-4-argon\", \"input\": \"Hello\", \"generation_config\": {\"thinking_level\": \"medium\"}}'\n```\n\nFor responses longer than 64K tokens, set `\"background\": true` and `max_output_tokens` above 65,536 (see the\n[long decode guide](https://ai.google.dev/gemini-api/docs/long-decode)). Remove `temperature`, `top_p`, `top_k` and explicit\n`cached_content` when migrating from Gemini 3.1 Pro ([migration guide](https://ai.google.dev/gemini-api/docs/latest-model)).\n\nBenchmark highlights (Google's table in the [model card](https://deepmind.google/models/model-cards/gemini-4-argon/), with methodology in\n[this PDF](https://storage.googleapis.com/deepmind-media/gemini/gemini_4_argon_model_evaluation.pdf)). Scores are for the highest\nthinking setting, and rival scores are mostly the providers' own figures.\n\n| Benchmark | Argon | GPT-6 Astra | Fable 5.1 | Opus 5.5 |\n|---|---|---|---|---|\n| Vals Index | 68.9 | 63.1 | 65.8 | 67.0 |\n| DeepSWE v1.1 | 77.9 | 74.1 | 67.4 | 74.2 |\n| FrontierSWE v2 | 55.0 | 65.5 | 56.3 | 62.3 |\n| Terminal-Bench 4.0 | 57.4 | 58.2 | 57.9 | 66.4 |\n| Terminal-Bench Science 0.1 | 57.6 | 68.1 | 52.6 | 63.3 |\n| GraphWalks 256K–1M | 84.2 | 71.8 | 65.0 | 66.8 |\n| OSWorld-2.0 (offline, partial) | 69.2 | 72.6 | — | — |\n| LVBench | 91.7 | 87.5 | 79.7 | 83.7 |\n| CWE-bench v1 | 68.0 | 68.0 | 58.0 | 67.0 |","page_url":"https://postcutoff.com/m/gemini-4-argon/","events_after":1918,"major_after":439,"historic_after":85,"missing_at_launch":1122,"events_since_release":null,"you_url":"https://postcutoff.com/you/gemini-4-argon/","briefings":{"xs":"https://postcutoff.com/briefings/cutoff-2026-03-xs.md","s":"https://postcutoff.com/briefings/cutoff-2026-03-s.md","m":"https://postcutoff.com/briefings/cutoff-2026-03-m.md","full":"https://postcutoff.com/briefings/cutoff-2026-03.md"}}