Compare / head-to-head
DeepSeek R1vs
GPT-4.1
DeepSeek R1 leads 6 of 8 shared benchmarks. GPT-4.1 has the larger context window (1.0M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
6 – 2
DeepSeek R1 leads
Cheaper per token
—
list price, input + output
Larger context
GPT-4.1
1.0M 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-4.1| Benchmark | DeepSeek R1 | GPT-4.1 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 8 | ||||
| SWE-bench Verified | 49.2% ↗ | 54.6% ↗ | -5.4 pt | GPT-4.1 |
| GPQA Diamond | 71.5% ↗ | 66.3% ↗ | +5.2 pt | DeepSeek R1 |
| MMLU | 90.8% ↗ | 90.2% ↗ | +0.6 pt | DeepSeek R1 |
| LMArena Elo | 1372.6 ↗ | 1414.8 ↗ | -42.2 | GPT-4.1 |
| MATH Level 5 (Epoch AI run) | 96.6% ↗epoch run | 83% ↗epoch run | +13.6 pt | DeepSeek R1 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 53.3% ↗epoch run | 38.3% ↗epoch run | +15 pt | DeepSeek R1 |
| SimpleBench (SimpleBench) | 30.9% ↗ | 27% ↗ | +3.9 pt | DeepSeek R1 |
| WeirdML (Håvard Tveit Ihle) | 41.6% ↗ | 39% ↗ | +2.6 pt | DeepSeek R1 |
| Only DeepSeek R1 reports · 3 | ||||
| MMLU-Pro | 84% ↗ | not reported | — | — |
| AIME 2025 | 70% ↗matharena | not reported | — | — |
| BALROG (BALROG) | 34.9% ↗ | not reported | — | — |
| Only GPT-4.1 reports · 7 | ||||
| AIME 2024 | not reported | 48.1% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 6% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 6% ↗epoch run | — | — |
| LMArena Vision (LMArena) | not reported | 1214 ↗ | — | — |
| SimpleQA Verified | not reported | 31.1% ↗epoch run | — | — |
| SWE-bench Verified (Epoch AI run) | not reported | 48.5% ↗epoch run | — | — |
| Video-MME (long, no subtitles) | not reported | 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 R1 minus GPT-4.1 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-4.1 | Edge |
|---|---|---|---|
| Context window | 128K tokens | 1.0M tokens | GPT-4.1 |
| Max output | 33K tokens | 33K tokens | Tie |
| Input price / 1M | — | $2 ↗ | — |
| Output price / 1M | — | $8 ↗ | — |
| Cached input / 1M | — | $0.5 ↗ | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | $10 | — |
| Modalities | text | text · vision | GPT-4.1 |
| Released | 2025-01-20 | 2025-04-14 | — |
| Cited benchmark scores | 15 | 21 | — |
Reliability
Provider status
Live from /statusMore matchups
DeepSeek R1 vs …
Models sharing the most benchmarksMore matchups
GPT-4.1 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-4.1 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.