Compare / head-to-head
DeepSeek R1vs
GPT-5
GPT-5 leads 8 of 9 shared benchmarks. GPT-5 has the larger context window (400K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 8
GPT-5 leads
Cheaper per token
—
list price, input + output
Larger context
GPT-5
400K 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
9 shared · 2 only DeepSeek R1 · 11 only GPT-5| Benchmark | DeepSeek R1 | GPT-5 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 9 | ||||
| SWE-bench Verified | 49.2% ↗ | 74.9% ↗ | -25.7 pt | GPT-5 |
| GPQA Diamond | 71.5% ↗ | 86.2% ↗epoch run | -14.7 pt | GPT-5 |
| AIME 2025 | 70% ↗matharena | 94.6% ↗ | -24.6 pt | GPT-5 |
| LMArena Elo | 1372.6 ↗ | 1434.5 ↗ | -61.9 | GPT-5 |
| BALROG (BALROG) | 34.9% ↗ | 32.8% ↗ | +2.1 pt | DeepSeek R1 |
| MATH Level 5 (Epoch AI run) | 96.6% ↗epoch run | 98.1% ↗epoch run | -1.5 pt | GPT-5 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 53.3% ↗epoch run | 91.4% ↗epoch run | -38.1 pt | GPT-5 |
| SimpleBench (SimpleBench) | 30.9% ↗ | 56.7% ↗ | -25.8 pt | GPT-5 |
| WeirdML (Håvard Tveit Ihle) | 41.6% ↗ | 60.7% ↗ | -19.1 pt | GPT-5 |
| Only DeepSeek R1 reports · 2 | ||||
| MMLU | 90.8% ↗ | not reported | — | — |
| MMLU-Pro | 84% ↗ | not reported | — | — |
| Only GPT-5 reports · 11 | ||||
| Chess Puzzles (Epoch AI run) | not reported | 37% ↗epoch run | — | — |
| EBR-bench (Epoch AI run) | not reported | 12.7% ↗epoch run | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | not reported | 22% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 55.4% ↗epoch run | — | — |
| GSO Opt@1 (GSO) | not reported | 5.9% ↗ | — | — |
| LMArena Vision (LMArena) | not reported | 1210.2 ↗ | — | — |
| LMArena WebDev (LMArena) | not reported | 1417.1 ↗ | — | — |
| Mystery Game Puzzles (Epoch AI run) | not reported | 23% ↗epoch run | — | — |
| SAGE (Vals AI) | not reported | 43.7% ↗ | — | — |
| SimpleQA Verified | not reported | 50.1% ↗epoch run | — | — |
| SWE-bench Verified (Epoch AI run) | not reported | 73.6% ↗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-5 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-5 | Edge |
|---|---|---|---|
| Context window | 128K tokens | 400K tokens | GPT-5 |
| Max output | 33K tokens | 128K tokens | GPT-5 |
| Input price / 1M | — | $1.25 ↗ | — |
| Output price / 1M | — | $10 ↗ | — |
| Cached input / 1M | — | $0.125 ↗ | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | $11.25 | — |
| Modalities | text | text · vision | GPT-5 |
| Released | 2025-01-20 | 2025-08-07 | — |
| Cited benchmark scores | 15 | 27 | — |
Reliability
Provider status
Live from /statusMore matchups
DeepSeek R1 vs …
Models sharing the most benchmarksMore matchups
GPT-5 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-5 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.