GPQA DiamondVerified
81.3%
GPQA Diamond (AA run)
Models / DeepSeek
DeepSeek R1
DeepSeek R1 · Released 2025-01-20
DeepSeek's January 2025 reasoning model. Open weights, chain-of-thought style.
Self-hosted reasoning route before V4 shipped. Still relevant for reproducible CoT research and local inference stacks.
Composite42
Context128K
Input / 1M$0.55
Output / 1M$2.2
Knowledge cutoffopen
Output speed31 t/s
Statuslightweight
Family profile
Best published score in each covered benchmark family.
51/100 avg
- Reasoning
- 81
- Agentic
- 16
- Long context
- 56
3 tested benchmark families
Benchmark placements
Per-benchmark positions
Where DeepSeek R1 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | Reasoning | #131/ 184 | 81 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #143/ 195 | Artificial Analysis | 2026-10-12 | |
| Humanity's Last Exam | Reasoning | #143/ 186 | 16 | Artificial Analysis | 2026-09-28 |
| Terminal-Bench | Agentic | #152/ 167 | 16 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #155/ 184 | 56 | Artificial Analysis | 2026-09-28 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where DeepSeek R1 shows up, it's highlighted.
ReasoningDeepSeek R1 · 81
- 196
- 295
- 395
Math
- 152
- 252
- 350
AgenticDeepSeek R1 · 16
- 191
- 289
- 389
Long contextDeepSeek R1 · 56
- 189
- 286
- 385
Newest receipts
5 recentLast 120 days
Benchmark rows added to the public ledger for DeepSeek R1 in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 31 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 55.67% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 81.3% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 15.8% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 15.91% | Artificial Analysis | 24 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where DeepSeek R1 sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for DeepSeek R1. Scores link to the original source; gaps mean no public row exists yet.
5 of 13 catalog benchmarks have a sourced row for DeepSeek R1.
Reasoning
Humanity's Last ExamVerified
15.8%
HLE (AA run)
Agentic
Terminal-BenchVerified
15.91%
Terminal-Bench Hard (AA run)
Long context
AA-LCRVerified
55.67%
AA-LCR (AA run)
Performance
AA output speedVerified
31 t/s
Median output tokens/s (1k prompt, default provider)
Sources
Where the numbers come from
- DeepSeek R1 model card· checked 2026-06-09
- DeepSeek R1 paper· checked 2026-06-09
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is DeepSeek R1?
DeepSeek R1 is a DeepSeek model: DeepSeek's January 2025 reasoning model. Open weights, chain-of-thought style.
What is DeepSeek R1 for?
Self-hosted reasoning route before V4 shipped. Still relevant for reproducible CoT research and local inference stacks.
What does DeepSeek R1 cost?
DeepSeek R1 lists $0.55 input and $2.2 output per 1M tokens. Pricing last checked 2026-06-09.
How large is the DeepSeek R1 context window?
DeepSeek R1 publishes a 128K token context window.
What is the VerdictPal status of DeepSeek R1?
DeepSeek R1 carries Lightweight status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.