DeepSeek R1 0528
100 cheaper ranked models score higher: GPT-6.1 Terra (65.4, $2.00), Muse Spark 1.1 (62.3, $2.00), Gemini 3.7 Flash (62.2, $1.50) and 97 more.
As of 8 Nov 2026, DeepSeek R1 0528 by DeepSeek scores 49.4 on the BenchLeader Index (593 ranked configurations), with its best results in human preference (60) and weakest in knowledge (34). API pricing: $1.85 per 1M input tokens and $2.75 per 1M output tokens ($2.08 blended at 3:1), list price on 8 Nov 2026. It streams about 26 tokens/s (Artificial Analysis). Released 28 May 2025.
Key facts (as of 8 Nov 2026)
- DeepSeek R1 0528 API pricing: $1.85 per 1M input tokens and $2.75 per 1M output tokens ($2.08 blended at 3:1) (list price, 8 Nov 2026).
- DeepSeek R1 0528 streams about 26 tokens/s (Artificial Analysis, 8 Nov 2026).
- DeepSeek R1 0528 was released on 28 May 2025 (DeepSeek).
In detail. DeepSeek R1 0528 is a DeepSeek open-weights reasoning model, released 28 May 2025. Its best configuration ranks #354 of 593 on the BenchLeader Index at 49.4 ±4.9, in the lower half. The ± is the point: 267 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in human preference (60) and lowest in knowledge (34). At $2.08 per million tokens blended it is pricier than most ranked models. Output speed of 26 tokens per second puts it in the slowest quarter, with a first answer in 86.0 s. Last measured 6 Nov 2026.
- Blended price
- $2.08/M
- $1.85 in · $2.75 out
- via Artificial Analysis
- Output speed
- 26 tok/s
- measured by Artificial Analysis
- First answer
- 86 s
- first token 1.92 s
- Context
- 164k
- via models.dev
- Full answer
- 144 s
- median, reasoning included
- Cost per task
- –
- Tokens per task
- –
- Released
- 28 May 2025
- Overall index49
- Reasoning44
- Coding59
- Agents & tools48
- Maths51
- Knowledge34
- Instruction following46
- Human preference60
- Long context52
Versions
DeepSeek has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| DeepSeek R1 0528this page | 28 May 2025 | 49.4 | #354 |
| DeepSeek R1 | 20 Jan 2025 | 47.3 | #401 |
How its index has moved
8 Sep 2026 to 8 Nov 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Metadata
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Openness Indexnot in index | 50#28 | Artificial Analysis |
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| GPQA Diamond | 76.3%#151 | Epoch AI Benchmarking Hub | |
| SimpleBench | 40.8%#59 | SimpleBench | |
| Arena Hard Prompts | 1433#134 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 81.3%#183 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 15.8%#259 | Artificial Analysis | |
| GPQA Diamond (HELM)not in index | 66.6%#15 | HELM Capabilities | |
| ARC-AGI-1 | 21.2%#223 | ARC Prize | |
| ARC-AGI-2 | 1.1%#230 | ARC Prize | |
| CritPtnot in index | 1.4%#222 | Artificial Analysis |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| WeirdML | 41.6%#96 | WeirdML | |
| Arena Coding | 1464#128 | Arena (LMArena) | |
| Aider Polyglot | 71.4%#10 | Aider polyglot leaderboard | |
| ALE-Benchnot in index | 804.1#77 | Epoch AI Benchmarking Hub | |
| Design Arena Websitenot in index | 1147#122 | Design Arena | |
| LiveCodeBench (AA)not in index | 77.0%#37 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 15.9%#194 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 36.5%#234 | Artificial Analysis | |
| DeepResearch Benchnot in index | 35.1%#39 | DeepResearch Bench | |
| METR Time Horizonsnot in index | 53.8%#24 | METR | |
| BrowseComp-Plusnot in index | 8.6%#18 | BrowseComp-Plus | |
| BrowseCompnot in index | 3.1%#8 | OpenAI |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| OTIS Mock AIME | 66.4%#162 | Epoch AI Benchmarking Hub | |
| MATH Level 5 | 96.6%#9 | Epoch AI Benchmarking Hub | |
| IMO 2025 | 6.8%#6 | MathArena | |
| MathArena Apex | 1.0%#36 | MathArena | |
| Omni-MATH (HELM) | 42.4%#26 | HELM Capabilities | |
| Arena Maths | 1392#163 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| PRBench Finance | 32.7%#38 | Scale AI SEAL | |
| PRBench Legal | 36.6%#40 | Scale AI SEAL | |
| MultiNRC | 27.6%#27 | Scale AI SEAL | |
| MMLU-Pro (HELM)not in index | 79.3%#21 | HELM Capabilities | |
| Global-MMLU-Lite (AA)not in index | 86.0%#26 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 39.6%#269 | Artificial Analysis | |
| IFEval (HELM) | 78.4%#51 | HELM Capabilities | |
| WildBench (HELM) | 82.8%#20 | HELM Capabilities | |
| Arena Instruction Followingnot in index | 1391#154 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1421#126 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1389#127 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1404#154 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1408#155 | Arena (LMArena) | |
| Lech Mazur Creative Writingnot in index | 81.9%#12 | Lech Mazur |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR v1.1 | 55.7%#308 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | 140.8#110 | Epoch AI Benchmarking Hub | |
| AA Intelligence Index v4.3.2not in index | 13.1#326 | Artificial Analysis | |
| AA Progress Index v1not in index | 230.8#316 | Artificial Analysis | |
| HELM Capabilities meannot in index | 69.9%#24 | HELM Capabilities |
Safety & honesty
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| MASKnot in index | 53.0%#46 | Scale AI SEAL |
API pricing: where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| StreamLake | 37 tok/s | 2.86 s | $0.571 | $2.29 | 128k | – |
| DeepInfra | 32 tok/s | 0.48 s | $0.500 | $2.15 | 164k | fp4 |
| SiliconFlow | 20 tok/s | 1.61 s | $0.500 | $2.18 | 164k | fp8 |
Price history
The price above, per 1M tokens, every day it changed.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.350 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0016 | $0.0011 | 1.6 min |
| Summarize a 30-page report | 12,000 / 600 | $0.024 | $0.010 | 1.8 min |
| Code edit | 6,000 / 1,500 | $0.015 | $0.0085 | 2.4 min |
| Agentic coding session | 60,000 / 4,000 | $0.122 | $0.054 | 4.0 min |
| Structured extraction | 2,000 / 200 | $0.0043 | $0.0020 | 1.6 min |
More from DeepSeek
See also
Embed or cite
Show where DeepSeek R1 0528 ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.
[](https://www.benchleader.com/models/deepseek-r1-0528)Cite this
The ranking changes daily, so a reference carries the data date. APA style:
BenchLeader. (2026, November 8). DeepSeek R1 0528 on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 8, 2026, from https://www.benchleader.com/models/deepseek-r1-0528
BibTeX:
@misc{benchleader_deepseek_r1_0528_20261108,
author = {{BenchLeader}},
title = {{DeepSeek R1 0528 on the BenchLeader Index (method v1.4)}},
year = {2026},
month = nov,
howpublished = {\url{https://www.benchleader.com/models/deepseek-r1-0528}},
note = {Data as of 2026-11-08}
}Every badge style, the embeddable top-10 widget and how to cite the Index
Questions people ask
- What are DeepSeek R1 0528's benchmark scores?
- As of 8 Nov 2026, DeepSeek R1 0528 scores 49.4 ±4.9 on the BenchLeader Index, strongest in human preference (60) and weakest in knowledge (34). That rests on independent benchmark results only, and every score on this page links to its publisher.
- How does DeepSeek R1 0528 rank?
- DeepSeek R1 0528 is #354 of 593 ranked configurations on 8 Nov 2026. 267 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 69.4.
- How much does DeepSeek R1 0528 cost per 1M tokens?
- As of 8 Nov 2026, DeepSeek R1 0528 costs $1.85 per 1M input tokens and $2.75 per 1M output tokens, $2.08 blended (three input tokens per output token). The list price last changed on 24 Oct 2026.
- When was DeepSeek R1 0528 released?
- DeepSeek released DeepSeek R1 0528 on 28 May 2025.
Data as of 8 Nov 2026. Compare with another model.
Cite as: BenchLeader, “DeepSeek R1 0528: benchmarks, pricing, speed and rank”, data as of 8 Nov 2026, https://www.benchleader.com/models/deepseek-r1-0528.