Gemini 4 Argon Benchmarks: How It Compares Against Opus 5.5 and GPT-6.1 Sol
Google has made Gemini 4 Argon broadly available, and on its model card it leads 13 of 19 benchmarks against GPT-6 Astra, Claude Fable 5.1 and Claude Opus 5.5. Opus 5.5 still wins Terminal-Bench 4.0 and PostTrainBench. Here are all the Gemini 4 Argon benchmarks, how it compares against Opus 5.5 and GPT-6.1 Sol, its pricing and where you can use it.
Category highlights
| Category | Gemini 4 Argon | GPT-6 Astra | Claude Opus 5.5 |
|---|---|---|---|
| Knowledge work: Vals Index | 68.9% | 63.1% | 67.0% |
| Finance | 65.4% | 53.5% | 58.6% |
| Legal | 19.6% | 5.4% | 3.8% |
| Agentic coding: DeepSWE | 77.9% | 74.1% | 74.2% |
| Terminal-Bench 4.0 | 57.4% | 58.2% | 66.4% |
| Science: LABBench 2 | 88.8% | 85.4% | 73.1% |
| Long context: 256K to 1M | 84.2% | 71.8% | 66.8% |
| Computer use: Agent's Last Exam | 39.5% | 34.2% | 38.2% |
| Visual reasoning: Chartography | 71.6% | 71.0% | 66.3% |
| Long-video understanding: LVBench | 91.7% | 87.5% | 83.7% |
All benchmarks
| Benchmark | Gemini 4 Argon | GPT-6 Astra | Claude Fable 5.1 | Claude Opus 5.5 |
|---|---|---|---|---|
| Vals Index | 68.9% | 63.1% | 65.8% | 67.0% |
| AutomationBench | 51.3% | 41.4% | 31.4% | 42.5% |
| Vals Finance Agent v2 | 65.4% | 53.5% | 58.9% | 58.6% |
| Harvey Legal Agent | 19.6% | 5.4% | 6.7% | 3.8% |
| DeepSWE v1.1 | 77.9% | 74.1% | 67.4% | 74.2% |
| FrontierSWE v2 | 55.0% | 65.5% | 56.3% | 62.3% |
| Vibe Code Bench | 91.9% | 89.6% | 90.3% | 90.3% |
| Terminal-Bench 4.0 | 57.4% | 58.2% | 57.9% | 66.4% |
| PostTrainBench | 45.3% | 44.3% | 40.2% | 49.3% |
| Terminal-Bench Science 0.1 | 57.6% | 68.1% | 52.6% | 63.3% |
| LABBench 2 | 88.8% | 85.4% | 68.6% | 73.1% |
| RiemannBench | 76.0% | 72.0% | 65.6% | 69.6% |
| GraphWalks up to 128K | 99.7% | 98.7% | 91.4% | 90.6% |
| GraphWalks 256K to 1M | 84.2% | 71.8% | 65.0% | 66.8% |
| Agent's Last Exam | 39.5% | 34.2% | - | 38.2% |
| OSWorld 2.0 Offline | 69.2% | 72.6% | - | - |
| Chartography | 71.6% | 71.0% | 46.2% | 66.3% |
| LVBench | 91.7% | 87.5% | 79.7% | 83.7% |
| CWE-bench v1 | 68.0% | 68.0% | 58.0% | 67.0% |
A dash means no score has been reported for that model.
Gemini 4 Argon vs Claude Opus 5.5 vs GPT-6.1 Sol
| Benchmark | Gemini 4 Argon | Claude Opus 5.5 | GPT-6.1 Sol |
|---|---|---|---|
| DeepSWE v1.1 | 77.9% | 74.2% | 75.2% |
| AutomationBench | 51.3% | 42.5% | 36.1% |
| Terminal-Bench Science 0.1 | 57.6% | 63.3% | 57.0% |
| OSWorld 2.0 Offline | 69.2% | - | 71.4% |
GPT-6.1 Sol has published scores on these four benchmarks. A dash means no score has been reported.
Winner by benchmark
| Benchmark | Winner |
|---|---|
| Vals Index | Gemini 4 Argon |
| AutomationBench | Gemini 4 Argon |
| Finance | Gemini 4 Argon |
| Legal | Gemini 4 Argon |
| DeepSWE | Gemini 4 Argon |
| FrontierSWE | GPT-6 Astra |
| Vibe Code Bench | Gemini 4 Argon |
| Terminal-Bench 4.0 | Claude Opus 5.5 |
| ML engineering: PostTrainBench | Claude Opus 5.5 |
| Terminal-Bench Science | GPT-6 Astra |
| LABBench 2 | Gemini 4 Argon |
| RiemannBench | Gemini 4 Argon |
| Long context up to 128K | Gemini 4 Argon |
| Long context 256K to 1M | Gemini 4 Argon |
| Agent's Last Exam | Gemini 4 Argon |
| OSWorld 2.0 | GPT-6 Astra |
| Chartography | Gemini 4 Argon |
| LVBench | Gemini 4 Argon |
| CWE-bench v1 | Gemini 4 Argon and GPT-6 Astra, tied |
API pricing
| Model | Input, per 1M tokens | Cached input | Output, per 1M tokens |
|---|---|---|---|
| Gemini 4 Argon, until November 30, 2026 | $2 | $0.10 | $10 |
| Gemini 4 Argon, from December 1, 2026 | $4 | $0.20 | $20 |
| Claude Opus 5.5 | $4 | $0.20 | $20 |
| GPT-6.1 Sol | $2 | $0.10 | $10 |
Specifications
| Specification | Gemini 4 Argon |
|---|---|
| Model ID | gemini-4-argon |
| Architecture | Sparse Mixture-of-Experts Transformer |
| Context window | 1M tokens |
| Maximum output | Up to 1M tokens |
| Knowledge cutoff | March 2026 |
| Text input | Yes |
| Image input | Yes |
| Audio input | Yes |
| Video input | Yes |
| Output | Text |
| Reasoning | Yes |
| Thinking levels | Low, Medium, High |
| Default thinking | Medium |
| Explicit context caching | No |
| Implicit caching | Yes |
| Function calling | Yes |
| Long-running background tasks | Yes |
| Release | October 13, 2026 |
Availability
| Platform | Gemini 4 Argon |
|---|---|
| Gemini API | Paid tier |
| Free API tier | No |
| Google AI Studio | Yes |
| Gemini app | Google AI Pro and Ultra |
| AI Mode in Search | Google AI Pro and Ultra |
| Google Antigravity | Yes |
| Enterprise | Yes |