Gemini 3.8 Flash vs Gemini 4 Argon
Vals Science Index?
Gemini 3.8 Flash35.8
Gemini 4 Argon55.4
| Benchmark | Gemini 3.8 Flash | Gemini 4 Argon |
|---|---|---|
| Humanity's Last Exam | 42.1 | 57.1 |
| Arena Text · overall | 1,496 · high | 1,524 |
| Arena Text · Chinese | 1,550.8 · high | 1,608.3 |
| Arena Text · English | 1,499 · high | 1,537.7 |
| Arena Text · instruction following | 1,491 · high | 1,535.6 |
| Arena Text · non-English | 1,491.1 · high | 1,521.7 |
| Arena Text · Russian | 1,514.4 · high | 1,534 |
| Arena Text · Spanish | 1,485.7 · high | 1,504.4 |
| AA Omniscience Index | 28.6 | 42.4 |
| SimpleQA Verified v1 | 74.6 | 76 |
| CritPt (physics) | 12.3 | 27.1 |
| AA Intelligence Index 4.3.2 | 39.8 | 52.6 |
| AA Omniscience · Accuracy | 53 | 49.9 |
| AA Omniscience · Hallucination Rate | 51.9 | 15.1 |
| AA-LCR (long context) 1.1 | 84 | 79.7 |
| GDP.pdf · AA all-pass | 22.8 | 21.8 |
| ARC-AGI-1 · public evaluation | 97.8 | 99.5 |
| ARC-AGI-1 · semi-private | 97.5 | 98.5 |
| ARC-AGI-2 · public evaluation | 87.5 | 97.1 |
| ARC-AGI-2 · semi-private | 82.9 | 91.3 |
| ARC-AGI-3 · semi-private | 4 | 49.8 · low |
| Chess Puzzles | 58 · high | 59 |
| FrontierMath Tier 4 (v2) | 22 · high | 78 |
| FrontierMath Tiers 1–3 (v2) | 68.4 · high | 86 |
| Mystery Game Puzzles | 47 · high | 64 |
| MathArena · ArXivMath · August 2026 | 40.4 | 82.5 |
| MathArena · ArXivMath · September 2026 | 50 | 87.7 |
| MathArena · BrokenArXiv · August 2026 | 27.4 | 54.2 |
| MathArena · BrokenArXiv · September 2026 | 26.7 | 49.7 |
| Vals LegalBench | 87 | 88.3 |
| Vals MedCode | 48.1 | 58.8 |
| Vals MedScribe | 84.5 | 87.4 |
| Vals Legal Research Bench | 38.9 | 54.8 |
| Vals Science Index | 35.8 | 55.4 |
| Arena Text · French | 1,491.7 · high | — |
| Arena Text · German | 1,497.6 · high | — |
| Arena Text · Japanese | 1,506.3 · high | — |
| Arena Text · Korean | 1,461.6 · high | — |
| Arena Text · Polish | 1,498.5 · high | — |
| MLCR-AA | 21.7 · high | — |
| FrontierMath Erdos | 0 · high | — |
| MathArena · ArXivMath · April 2026 | 64.2 | — |
| MathArena · ArXivMath · June 2026 | 67.4 | — |
| MathArena · ArXivMath · May 2026 | 58.3 | — |
| MathArena · BrokenArXiv · April 2026 | 33.6 | — |
| MathArena · BrokenArXiv · June 2026 | 20.8 | — |
| MathArena · BrokenArXiv · May 2026 | 19 | — |
| MathArena · ArXivLean · June 2026 | 19.6 | — |
| MathArena · ArXivLean · September 2026 | — | 39.6 |
Specs
| Spec | Gemini 3.8 Flash | Gemini 4 Argon |
|---|---|---|
| Input, per 1M | !$0.75 | !$2 |
| Output, per 1M | !$3.75 | !$10 |
| Cost per task? | $0.93 | $1.99 |
| Speed, tok/s | — | 84 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | Sep 2026 | Oct 2026 |
| See images | ✓ | ✓ |
| Read PDFs | ✓ | ✓ |
| Hear audio | ✓ | ✓ |
| Watch video | ✓ | ✓ |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ✓ | ✓ |
| Run code | ✓ | ✓ |
| Use a computer | ? | ? |
What breaks when you switch
Gemini 3.8 Flash → Gemini 4 Argon
Synchronous output stops at 65,536›
With background false, output limits above 65,536 tokens return HTTP 400; stay at or below that limit.
Sampling parameters are ignored›
temperature, top_p and top_k are ignored; remove them from requests.
No explicit caching›
Explicit caching is unavailable; omit explicit cache requests.
Gemini 4 Argon → Gemini 3.8 Flash
Sampling parameters are ignored›
temperature, top_p and top_k are ignored now and will cause HTTP 400 in future generations; remove them from requests.
No minimal thinking level›
The minimal thinking level returns an error; avoid minimal.