Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
DeepSeek · Open Weight · bench-align-v5
DeepSeek-R1-0528 is a general-purpose language model. It accepts text and produces text.
DeepSeek-R1-0528 by DeepSeek costs $0.57 per 1M input tokens and $2.29 per 1M output tokens and carries a context window of 128K tokens — observed 2026-11-21.
Seven axes from the ranking source. A missing axis is drawn as a gap.
Each axis shows the ranking source’s score and, beside it, BenchLM’s separate verified score for that axis. The two are published side by side and are not interchangeable.
Every published price component, including cache reads and writes.
10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.
Lifecycle events the source associates with this model, plus what this profile does not claim.
That is not evidence the model has no lifecycle plan — only that this source published none.
Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Evidence key: ObservedUnavailable
| Measurement | Value | Observed | Last good | Evidence |
|---|---|---|---|---|
| Time to first token | Unavailable | Unavailable | Unavailable | |
| Throughput | 27 tok/s | 2026-11-21 | 2026-11-21 |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| GPQA Graduate-Level Google-Proof Q&A | 81 | #69 | 448 questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| MMLU-Pro Massive Multitask Language Understanding Professional | 85 | #21 | Multiple subjects | Professional level | MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark |
| HLE Humanity's Last Exam | 17.7 | #73 | Expert-level questions | Frontier expert level | Humanity's Last Exam |
| SimpleQA Measuring Short-Form Factuality in Large Language Models | 27.8 | #10 | Factual questions | Factual accuracy focused | Measuring short-form factuality in large language models |
| MMLU-Redux MMLU-Redux | 93.4 | #8 | Broad academic QA | Advanced general knowledge | Qwen3.6 launch benchmarks |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Codeforces Codeforces Rating | 1930.0 | #6 | Competitive programming contests | Elite competitive programming | DeepSeek-V4 Technical Report |
| SWE-bench Verified Software Engineering Benchmark Verified | 57.6 | #92 | 500 verified issues | Professional software engineering | SWE-bench: Can Language Models Resolve Real-World GitHub Issues? |
| LiveCodeBench LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code | 73.3 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AIME 2024 American Invitational Mathematics Examination 2024 | 91.4 | #6 | 15 problems | High school olympiad level | American Invitational Mathematics Examination |
| AIME 2025 American Invitational Mathematics Examination 2025 | 87.5 | #23 | 15 problems | High school olympiad level | American Invitational Mathematics Examination |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Design Arena Website Design Arena Website Elo | 1137 | #91 | Website generation comparisons | Design and website generation | OpenRouter Grok 4.3 benchmarks |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| FailureSensorIQ FailureSensorIQ | 29.1 | #22 | Multiple-choice questions on sensors and failure modes for 10 industrial assets | Industrial maintenance and reliability knowledge | BenchLM |
A second, independent score source. Epoch’s own figures in Epoch’s units, one line per model version Epoch evaluated (effort setting as Epoch labels it). They are not part of this site’s capability scores or rank.
| Benchmark | Variant | Score | Epoch date | Evidence |
|---|---|---|---|---|
| GPQA Diamond Accuracy on graduate-level science multiple-choice questions (best score across scorers). Unit: fraction (0–1); higher is better. | default | 0.7633 ± 0.024 | 2025-05-29 | |
| Aider polyglot Percent of coding exercises solved, as compiled by Epoch from Aider's leaderboard. Unit: percent; higher is better. | default diff | 71.4% | 2025-06-06 | |
| METR time horizon Length of software task (in human minutes) the model completes with 50% reliability, as compiled by Epoch from METR. Unit: minutes; higher is better. | default METR version not stated | 31.2 min | — |
Source: Epoch AI, CC BY 4.0 (licence), benchmark data read once a day.
| #9 |
| Continuously updated contest problems |
| Competitive programming level |
| LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code |