Gemini 4 Argon
| Released | 2026-10-13 |
|---|---|
| Price per 1M | $2.00 in · $10.00 out |
| Overall score | 64.8 · rank 7 |
| Context window | 1,048,576 tokens |
| Provider | |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 64.8 | 7 | 56 |
| Overall: Coding | 64.1 | 9 | 42 |
| Overall: Reasoning | 66.9 | 5 | 21 |
| Overall: Agentic work | 64.6 | 5 | 39 |
| Overall: Frontend | 61.5 | 12 | 14 |
| Coding | 64.4 | 8 | 11 |
| Agents | 63.7 | 6 | 15 |
| Reasoning | 68.8 | 4 | 13 |
| Math | 64.9 | 11 | 6 |
| Knowledge | 66.1 | 3 | 2 |
| Frontend | 60.1 | 19 | 3 |
| General | 62.7 | 14 | 6 |
| Value vs price | +12.9 | 15 |
Buzz
| Posts on X | About 240 a day |
|---|---|
| Mood | 16% positive, 16% negative |
| Hacker News | 15 mentions this week |
| Checked | 2026-11-11 |
- Gemini 4 Carbon (high) takes #1 on the Extended NYT Connections Benchmark, scoring 99.3. Gemini 4 Argon (high), the previous leader, scored 98.6. It fully solves 927 of 940 puzzles. More info: https://t.co/R7xOCBqxgw https://t.co/kb2Ox4fOae
- gemini 4 carbon is live coded with Gemini 4 Argon once and it was a beast now I'm on Gemini 4 Carbon and it's really good too. Easy on tokens as well everyone laughed at Google when they were in last place. now they're second
- 🚀 25 AI tools everyone is talking about in November 2026. Ranked by their buzz on X. Whether you're a developer, designer, content creator, or someone exploring AI, here are 25 tools worth knowing: 1. Claude – Long-context writing and thoughtful coding assistance. 2. ChatGPT…
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 67.1 ±1.9 | 17 |
| High | 64.9 ±0.9 | 39 |
| Low | 63.1 ±4.5 | 4 |
| Medium | 65.3 ±4.1 | 5 |
| Minimal | 63.2 ±5.0 | 1 |
Evidence
Coding 64.4 ±1.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CursorBench | 52.9% | Epoch AI | 2026-11-11 |
| FrontierSWE | 55.0% | Epoch AI | 2026-11-11 |
| FrontierCode | 47.6% | Epoch AI | 2026-11-11 |
| Code Migration v1.1 | 70.8% | Vals AI | 2026-11-09 |
| LiveBench 2.0 Coding | 85.7% | LiveBench | 2026-11-06 |
| WeirdML v3 | 20.3% | Epoch AI | 2026-11-11 |
| MirrorCode | 58.9% | Epoch AI | 2026-10-14 |
| IOI | 100.0% | Vals AI | 2026-10-29 |
| SWE-bench Multimodal | 56.0% | Vals AI | 2026-11-09 |
| SciCode | 61.8% | Epoch AI | 2026-11-11 |
| ProgramBench | 2.5% | Vals AI | 2026-11-09 |
Agents 63.7 ±1.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4.0 (Vals run) | 57.6% | Vals AI | 2026-11-09 |
| SRE Bench | 44.3% | Vals AI | 2026-11-05 |
| Terminal-Bench-Science | 44.3% | Vals AI | 2026-11-09 |
| APEX-Agents | 82.4% | Epoch AI | 2026-11-11 |
| Remote Labor Index | 17.1% | Epoch AI | 2026-11-11 |
| LiveBench 2.0 Agentic | 58.5% | LiveBench | 2026-11-06 |
| LiveBench 2.0 Browser Use | 82.8% | LiveBench | 2026-11-06 |
| Terminal-Bench 2.1 (Vals run) | 57.7% | Vals AI | 2026-10-01 |
| EMB v1.1 (Excel financial modeling) | 74.8% | Vals AI | 2026-11-09 |
| LiveBench 2.0 Research | 86.8% | LiveBench | 2026-11-06 |
| Tax Agent Bench | 49.2% | Vals AI | 2026-11-09 |
| Finance Agent v2.1 | 67.7% | Vals AI | 2026-11-09 |
| Vending-Bench 2 | $13,718 | Epoch AI | 2026-11-11 |
| HLAB-Vals 1.1 (legal agent) | 27.1% | Vals AI | 2026-11-09 |
| CyberBench v1.1 | 77.9% | Vals AI | 2026-11-05 |
Reasoning 68.8 ±1.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 91.3% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 91.3% | ARC Prize | 2026-11-11 |
| Extended NYT Connections | 98.6% | Lech Mazur | 2026-11-10 |
| MysteryMechanism | 45.5% | Vals AI | 2026-11-09 |
| GPQA Diamond | 96.4% | Epoch AI | 2026-10-13 |
| CritPt | 27.1% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 98.5% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 98.5% | Epoch AI | 2026-11-11 |
| SimpleBench | 91.7% | SimpleBench | 2026-11-10 |
| BioMysteryBench | 76.3% | Vals AI | 2026-11-09 |
| ARC-AGI-3 | 97.5% | ARC Prize | 2026-11-11 |
| Chess Puzzles | 59.0% | Epoch AI | 2026-10-13 |
| LiveBench 2.0 Reasoning | 95.1% | LiveBench | 2026-11-06 |
| Mystery Game Puzzles | 64.0% | Epoch AI | 2026-10-13 |
| LiveBench 2.0 Data Analysis | 85.2% | LiveBench | 2026-11-06 |
Math 64.9 ±1.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 86.0% | Epoch AI | 2026-10-13 |
| FrontierMath Tier 4 | 78.1% | Epoch AI | 2026-10-13 |
| MathArena ArXivMath (latest month) | 83.0% | MathArena | 2026-11-11 |
| OTIS Mock AIME 2024-2025 | 100.0% | Epoch AI | 2026-10-13 |
| ProofBench v1.1 | 99.0% | Vals AI | 2026-11-09 |
| ProofBench | 99.0% | Epoch AI | 2026-11-11 |
| MathArena BrokenArXiv (latest month) | 39.8% | MathArena | 2026-11-11 |
Knowledge 66.1 ±2.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 76.0% | Epoch AI | 2026-10-13 |
| Legal Research Bench | 54.8% | Vals AI | 2026-11-09 |
Frontend 60.1 ±5.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LiveBench 2.0 Design | 69.2% | LiveBench | 2026-11-06 |
| Vibe Code Bench v1.1 | 91.9% | Vals AI | 2026-11-09 |
| LMArena WebDev | 1,683 | LMArena | 2026-11-09 |
General 62.7 ±1.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LLM Debate Benchmark | 1,669 | Lech Mazur | 2026-10-26 |
| EQ-Bench 4 | 1,236 | EQ-Bench | 2026-10-24 |
| EQ-Bench Creative Writing v3 | 1,758 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,522 | LMArena | 2026-11-09 |
| Creative Story-Writing (Lech Mazur) | 61.0% | Lech Mazur | 2026-10-23 |
| EQ-Bench Longform Creative Writing | 80.9% | EQ-Bench | 2026-11-11 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-1 | 98.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 76.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 98.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 76.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 | 87.1% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 42.1% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 87.1% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 42.1% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 97.3% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 46.3% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 97.3% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 49.8% | ARC Prize | A stronger configuration result was used |
| CursorBench | 52.8% | Epoch AI | A stronger configuration result was used |
| CursorBench | 50.6% | Epoch AI | A stronger configuration result was used |
| CursorBench | 48.8% | Epoch AI | A stronger configuration result was used |
Names matched to this model
gemini-4-argon