| Output token limit | 1M tokens (up from 64K) |
|---|
| Introductory price | $2 per 1M input tokens, $10 per 1M output tokens |
|---|
| Cached input price | 95% off the input token price |
|---|
| DeepSWE v1.1 | 77.9% (Google: new state of the art) |
|---|
| AutomationBench (Zapier) | 51.3%, ranked #1 |
|---|
| LVBench (long video) | 91.7%, state of the art |
|---|
| CWE-bench v1 | 68%, tied for first |
|---|
All figures are Google's own claims from the announcement; the page gives no rival scores in text, so no comparison graph is saved. · source
Estimate from the version's published specs, benchmarks and reviews, 0-5; scores are only comparable within one criteria year. Maker's 30 Sep 2026 post: 1M output-token limit (up from 64K), DeepSWE v1.1 77.9% (Google-run harness), LVBench 91.7%, CWE-bench v1 68.0% tied first; $2/$10 per M tokens intro then $4/$20. eesel (citing Artificial Analysis) gives Intelligence Index 53 vs Opus 5.5 at 58, Terminal-Bench 4.0 57.1% (4th), cost $1.99 promo / $3.98 per task, 62K output tokens per task. Speed not published (AA lists N/A); not API-callable at review time. Input context window not stated on the opened page, so context reflects the 1M output limit and GraphWalks 84.2% at 256K-1M. Only two independent write-ups opened; no hands-on tests found.[1]
Gemini 4 Argon is the best model Google has shipped, and on paper it lands in the same tier as GPT-6 Astra.
Independent evaluations indicate that while Argon scores competitively on several key intelligence benchmarks, it shares the lead rather than achieving a clear, universal advantage.
| Gemini 4 Argon review: a great model you can't use yet · Oct 1, 2026 | 0.70 |
|---|---|
| Google Gemini 4 Argon closes the gap with OpenAI and Anthropic, but doesn’t take a clear lead · Oct 1, 2026 | 0.20 |
Photos from readers, kept apart from the pin's sourced media. Rate what you ate or saw, and on a restaurant pin tag the dishes right on your photo.
The first entry is always the pin's source. Overall confidence is a weighted average of how firmly each reference supports the start and end times used above; a reference counts half as much for every 180 days older than the newest.
Google’s September 30 announcement introduced Gemini 4 Argon to trusted cyber defenders. Broad paying-user release followed October 13 and is covered by pin 15308, which carries the canonical release rubric. Pin 4810 is an announcement duplicate, not the release rubric.
DeepMind's Gemini page now leads with Gemini 4 Argon and carries its benchmark table against GPT-6 Astra, Claude Fable 5.1 and Claude Opus 5.5, plus the four safeguard areas Google is strengthening before broad availability.
Something missing or wrong? Say it in your own words: a link that backs this pin up, a different start or end date and why, or a fact it lacks or gets wrong. The AI checks it against this pin's sources, searches for better ones, and adds any page that backs you up. The pin's own sources still count most. A picture that shows something else, or shows it badly, is looked at too, and moved down or replaced.
| Benchmark | Gemini 4 Argon | GPT-6 Astra | Claude Fable 5.1 | Claude Opus 5.5 |
|---|---|---|---|---|
| Vals Index (Knowledge work) | 68.9% | 63.1% | 65.8% | 67.0% |
| AutomationBench (Knowledge work, Score) | 51.3% | 41.4% | 31.4% | 42.5% |
| Vals Finance Agent v2 (Knowledge work) | 65.4% | 53.5% | 58.9% | 58.6% |
| Harvey's Legal Agent Benchmark (Knowledge work) | 19.6% | 5.4% | 6.7% | 3.8% |
| DeepSWE v1.1 (Agentic coding) | 77.9% | 74.1% | 67.4% | 74.2% |
| FrontierSWE v2 (Agentic coding) | 55.0% | 65.5% | 56.3% | 62.3% |
| Vibe Code Bench (Agentic coding) | 91.9% | 89.6% | 90.3% | 90.3% |
| Terminal-bench 4.0 (Agentic coding) | 57.4% | 58.2% | 57.9% | 66.4% |
| PostTrainBench (ML engineering) | 45.3% | 44.3% | 40.2% | 49.3% |
| Terminal-Bench Science 0.1 (Science and math) | 57.6% | 68.1% | 52.6% | 63.3% |
| LABBench 2 (Science and math) | 88.8% | 85.4% | 68.6% | 73.1% |
| RiemannBench (Science and math) | 76.0% | 72.0% | 65.6% | 69.6% |
| GraphWalks (Long context, Up to 128k, BFS (F1)) | 99.7% | 98.7% | 91.4% | 90.6% |
| GraphWalks (Long context, 256k to 1M, BFS (F1)) | 84.2% | 71.8% | 65.0% | 66.8% |
| Agent's Last Exam (Computer use, Pass rate) | 39.5% | 34.2% | — | 38.2% |
| OSWorld-2.0 (Computer use, Offline subsetPartial score) | 69.2% | 72.6% | — | — |
| Chartography (Multimodal understanding) | 71.6% | 71.0% | 46.2% | 66.3% |
| LVBench (Multimodal understanding) | 91.7% | 87.5% | 79.7% | 83.7% |
| CWE-bench v1 (Cybersecurity) | 68.0% | 68.0% | 58.0% | 67.0% |
The same event, pinned 1 more time. On the timeline they stack, the most watched (then most viewed) on top: this one.