Compare / head-to-head
Gemini 3.6 FlashvsGPT-5.4 Pro
GPT-5.4 Pro leads 3 of 4 shared benchmarks. Gemini 3.6 Flash is 46.7x cheaper per token. GPT-5.4 Pro has the larger context window (1.1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 3
GPT-5.4 Pro leads
Cheaper per token
Gemini 3.6 Flash
46.7x cheaper, input + output
Larger context
GPT-5.4 Pro
1.1M tokens
Provider uptime (30d)
100% · 94.2446%
Google · OpenAI
Gemini 3.6 Flash
Google · released 2026-07-21
textvisionaudio
GPT-5.4 Pro
OpenAI · released 2026-03-05
textvision
Quality
Benchmark matrix
4 shared · 10 only Gemini 3.6 Flash · 8 only GPT-5.4 Pro| Benchmark | Gemini 3.6 Flash | GPT-5.4 Pro | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 4 | ||||
| GPQA Diamond | 94.1% ↗epoch | 94.4% ↗ | -0.3 pt | GPT-5.4 Pro |
| FrontierMath Tier 4 v2 (Epoch AI run) | 22% ↗epoch | 58.5% ↗epoch | -36.5 pt | GPT-5.4 Pro |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 58.9% ↗epoch | 82.5% ↗epoch | -23.6 pt | GPT-5.4 Pro |
| SimpleQA Verified | 66.2% ↗epoch | 46.3% ↗epoch | +19.9 pt | Gemini 3.6 Flash |
| Only Gemini 3.6 Flash reports · 10 | ||||
| AIME 2026 | 96.7% ↗matharena ⚠ | not reported | — | — |
| CharXiv Reasoning (no tools) | 85.2% ↗ | not reported | — | — |
| DeepSWE v1.1 | 49% ↗ | not reported | — | — |
| GDM-MRCR v2 (128k average) | 91.8% ↗ | not reported | — | — |
| GDPVal-AA v2 | 1421 ↗ | not reported | — | — |
| MLE-Bench | 63.9% ↗ | not reported | — | — |
| OSWorld-Verified | 83% ↗ | not reported | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 94.2% ↗epoch | not reported | — | — |
| SWE-Bench Pro (Public) | 58.7% ↗ | not reported | — | — |
| Terminal-bench 2.1 | 78% ↗ | not reported | — | — |
| Only GPT-5.4 Pro reports · 8 | ||||
| Humanity's Last Exam (no tools) | not reported | 42.7% ↗ | — | — |
| ARC-AGI-1 (Verified) | not reported | 94.5% ↗ | — | — |
| ARC-AGI-2 (Verified) | not reported | 83.3% ↗ | — | — |
| BrowseComp | not reported | 89.3% ↗ | — | — |
| FrontierMath Tier 1-3 | not reported | 50% ↗ | — | — |
| FrontierMath Tier 4 | not reported | 38% ↗ | — | — |
| GDPval (wins or ties) | not reported | 82% ↗ | — | — |
| Humanity's Last Exam (with tools) | not reported | 58.7% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Gemini 3.6 Flash minus GPT-5.4 Pro in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Gemini 3.6 Flash | GPT-5.4 Pro | Edge |
|---|---|---|---|
| Context window | 1.0M tokens | 1.1M tokens | GPT-5.4 Pro |
| Max output | 66K tokens | 128K tokens | GPT-5.4 Pro |
| Input price / 1M | $0.75 ↗ | $30 ↗ | Gemini 3.6 Flash |
| Output price / 1M | $3.75 ↗ | $180 ↗ | Gemini 3.6 Flash |
| Cached input / 1M | $0.075 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $4.5 | $210 | Gemini 3.6 Flash |
| Modalities | text · vision · audio | text · vision | Gemini 3.6 Flash |
| Released | 2026-07-21 | 2026-03-05 | — |
| Cited benchmark scores | 14 | 13 | — |
Reliability
Provider status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
Gemini 3.6 Flash vs …
Models sharing the most benchmarksMore matchups
GPT-5.4 Pro vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Gemini 3.6 Flash and GPT-5.4 Pro on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.