Compare / head-to-head
DeepSeek V3.2-Specialevs
Gemini 3 Flash Preview
Gemini 3 Flash Preview leads 3 of 4 shared benchmarks.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 3
Gemini 3 Flash Preview leads
Cheaper per token
—
list price, input + output
Larger context
—
unknown
Providers
2 providers
DeepSeek · Google
DeepSeek V3.2-Speciale
DeepSeek · released 2025-12-01
text
- Context
- —
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 9 · 4 core
Gemini 3 Flash Preview
Google · released 2025-12-17
textvisionaudio
- Context
- 1.0M
- Max out
- 66K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 34 · 12 core
Quality
Benchmark matrix
4 shared · 5 only DeepSeek V3.2-Speciale · 22 only Gemini 3 Flash Preview| Benchmark | DeepSeek V3.2-Speciale | Gemini 3 Flash Preview | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 4 | ||||
| GPQA Diamond | 85.7% ↗ | 90.4% ↗ | -4.7 pt | Gemini 3 Flash Preview |
| AIME 2025 | 96% ↗ | 95.2% ↗ | +0.8 pt | DeepSeek V3.2-Speciale |
| SimpleBench (SimpleBench) | 52.6% ↗ | 61.1% ↗ | -8.5 pt | Gemini 3 Flash Preview |
| WeirdML (Håvard Tveit Ihle) | 46.7% ↗ | 61.6% ↗ | -14.9 pt | Gemini 3 Flash Preview |
| Only DeepSeek V3.2-Speciale reports · 5 | ||||
| HLE (text-only) | 30.6% ↗ | not reported | — | — |
| HMMT Feb 2025 | 99.2% ↗ | not reported | — | — |
| HMMT Nov 2025 | 94.4% ↗ | not reported | — | — |
| IMOAnswerBench | 84.5% ↗ | not reported | — | — |
| LiveCodeBench (Pass@1-COT) | 88.7% ↗ | not reported | — | — |
| Only Gemini 3 Flash Preview reports · 22 | ||||
| SWE-bench Verified | not reported | 78% ↗ | — | — |
| Humanity's Last Exam (no tools) | not reported | 33.7% ↗ | — | — |
| LMArena Elo | not reported | 1472.7 ↗ | — | — |
| BALROG (BALROG) | not reported | 48.1% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 40% ↗epoch run | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | not reported | 17.1% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 51.2% ↗epoch run | — | — |
| GSO Opt@1 (GSO) | not reported | 7.8% ↗ | — | — |
| LMArena Vision (LMArena) | not reported | 1272.9 ↗ | — | — |
| LMArena WebDev (LMArena) | not reported | 1438.9 ↗ | — | — |
| MMMU-Pro | not reported | 81.2% ↗ | — | — |
| Mystery Game Puzzles (Epoch AI run) | not reported | 26% ↗epoch run | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | not reported | 95.6% ↗epoch run | — | — |
| SAGE (Vals AI) | not reported | 51.8% ↗ | — | — |
| SimpleQA Verified | not reported | 68.7% ↗ | — | — |
| SWE-bench Verified (Epoch AI run) | not reported | 75.4% ↗epoch run | — | — |
| tau2-bench Airline (Sierra) | not reported | 82.5% ↗ | — | — |
| tau2-bench Banking Knowledge (Sierra) | not reported | 27.3% ↗ | — | — |
| tau2-bench Retail (Sierra) | not reported | 76.8% ↗ | — | — |
| tau2-bench Telecom (Sierra) | not reported | 91.2% ↗ | — | — |
| Terminal-Bench 2.0 | not reported | 47.6% ↗ | — | — |
| Vending-Bench 2 (Andon Labs) | not reported | 3634.72 ↗ | — | — |
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 DeepSeek V3.2-Speciale minus Gemini 3 Flash Preview 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 | DeepSeek V3.2-Speciale | Gemini 3 Flash Preview | Edge |
|---|---|---|---|
| Context window | — | 1.0M tokens | — |
| Max output | — | 66K tokens | — |
| Input price / 1M | — | — | — |
| Output price / 1M | — | — | — |
| Cached input / 1M | — | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | — | — |
| Modalities | text | text · vision · audio | Gemini 3 Flash Preview |
| Released | 2025-12-01 | 2025-12-17 | — |
| Cited benchmark scores | 9 | 34 | — |
Reliability
Provider status
Live from /statusMore matchups
DeepSeek V3.2-Speciale vs …
Models sharing the most benchmarksMore matchups
Gemini 3 Flash Preview vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run DeepSeek V3.2-Speciale and Gemini 3 Flash Preview on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.