Compare / head-to-head
DeepSeek R1vs
GPT-4o
DeepSeek R1 leads 8 of 8 shared benchmarks. Both offer a 128K-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
8 – 0
DeepSeek R1 leads
Cheaper per token
—
list price, input + output
Larger context
Tie
both 128K tokens
Providers
2 providers
DeepSeek · OpenAI
DeepSeek R1
DeepSeek · released 2025-01-20
text
- Context
- 128K
- Max out
- 33K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 15 · 7 core
Quality
Benchmark matrix
8 shared · 3 only DeepSeek R1 · 7 only GPT-4o| Benchmark | DeepSeek R1 | GPT-4o | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 8 | ||||
| GPQA Diamond | 71.5% ↗ | 46% ↗ | +25.5 pt | DeepSeek R1 |
| MMLU | 90.8% ↗ | 85.7% ↗ | +5.1 pt | DeepSeek R1 |
| AIME 2025 | 70% ↗matharena | 11.7% ↗matharena | +58.3 pt | DeepSeek R1 |
| LMArena Elo | 1372.6 ↗ | 1346.1 ↗ | +26.5 | DeepSeek R1 |
| BALROG (BALROG) | 34.9% ↗ | 32.3% ↗ | +2.6 pt | DeepSeek R1 |
| MATH Level 5 (Epoch AI run) | 96.6% ↗epoch run | 53.3% ↗epoch run | +43.3 pt | DeepSeek R1 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 53.3% ↗epoch run | 6.4% ↗epoch run | +46.9 pt | DeepSeek R1 |
| WeirdML (Håvard Tveit Ihle) | 41.6% ↗ | 25.1% ↗ | +16.5 pt | DeepSeek R1 |
| Only DeepSeek R1 reports · 3 | ||||
| SWE-bench Verified | 49.2% ↗ | not reported | — | — |
| MMLU-Pro | 84% ↗ | not reported | — | — |
| SimpleBench (SimpleBench) | 30.9% ↗ | not reported | — | — |
| Only GPT-4o reports · 7 | ||||
| AIME 2024 | not reported | 13.1% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 13% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 0.4% ↗epoch run | — | — |
| GSO Opt@1 (GSO) | not reported | 0% ↗ | — | — |
| LMArena Vision (LMArena) | not reported | 1161.8 ↗ | — | — |
| SimpleQA Verified | not reported | 26% ↗epoch run | — | — |
| SWE-bench Verified (Epoch AI run) | not reported | 31% ↗epoch run | — | — |
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 R1 minus GPT-4o 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 R1 | GPT-4o | Edge |
|---|---|---|---|
| Context window | 128K tokens | 128K tokens | Tie |
| Max output | 33K tokens | 16K tokens | DeepSeek R1 |
| Input price / 1M | — | $2.5 ↗ | — |
| Output price / 1M | — | $10 ↗ | — |
| Cached input / 1M | — | $1.25 ↗ | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | $12.5 | — |
| Modalities | text | text · vision | GPT-4o |
| Released | 2025-01-20 | 2024-05-13 | — |
| Cited benchmark scores | 15 | 21 | — |
Reliability
Provider status
Live from /statusMore matchups
DeepSeek R1 vs …
Models sharing the most benchmarksMore matchups
GPT-4o vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run DeepSeek R1 and GPT-4o on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.