The AI arena is free today

Open Superagent

DeepSeek-R1-0528 vs Step3-VL-10B

DeepSeek-R1-0528 and Step3-VL-10B are closely matched at 24.1 and 25.5 on the LLM Stats Score.

DeepSeek · StepFun · Updated for 2026

Which is better?

DeepSeek-R1-0528 and Step3-VL-10B are closely matched on the overall LLM Stats Score at 24.1 and 25.5.

In the 1 individual benchmarks reported for both models, Step3-VL-10B wins 1; this is a narrower head-to-head signal than the composite indexes.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose DeepSeek-R1-0528

  • you want predictable pricing at $0.50/M input and $2.15/M output

Choose Step3-VL-10B

  • you value its reported benchmark strengths — it wins 1 of 1 exact shared results
  • you want the most recent training data — it shipped Jan 2026

At a glance

The differences that matter most.

Core performance indexes
24.1
#166
25.5
#155
23.7
#162
25.4
#150
Cost, coverage & limits
Benchmark wins
0 of 1
1 of 1
Input price
$0.50 / M
— / M
Output price
$2.15 / M
— / M
Context window
163,840

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
DeepSeek-R1-0528
Step3-VL-10B
26.2#104
23.5#126
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

16 reported for DeepSeek-R1-0528 · 6 for Step3-VL-10B

1 shared

DeepSeek-R1-0528 outperforms in 0 benchmarks, while Step3-VL-10B is better at 1 benchmark (AIME 2025).

Step3-VL-10B significantly outperforms across most benchmarks.

Sun Sep 13 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

661.0B diff

DeepSeek-R1-0528 has 661.0B more parameters than Step3-VL-10B, making it 6610.0% larger.

DeepSeek
DeepSeek-R1-0528
671.0Bparameters
StepFun
Step3-VL-10B
10.0Bparameters
671.0B
DeepSeek-R1-0528
10.0B
Step3-VL-10B

Context Window

Maximum input and output token capacity

Only DeepSeek-R1-0528 specifies input context (163,840 tokens). Only DeepSeek-R1-0528 specifies output context (163,840 tokens).

DeepSeek
DeepSeek-R1-0528
Input163,840 tokens
Output163,840 tokens
StepFun
Step3-VL-10B
Input- tokens
Output- tokens
Sun Sep 13 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Step3-VL-10B supports multimodal inputs, whereas DeepSeek-R1-0528 does not.

Step3-VL-10B can handle both text and other forms of data like images, making it suitable for multimodal applications.

DeepSeek-R1-0528

Text
Images
Audio
Video

Step3-VL-10B

Text
Images
Audio
Video

License

Usage and distribution terms

DeepSeek-R1-0528 is licensed under MIT, while Step3-VL-10B uses Apache 2.0.

License differences may affect how you can use these models in commercial or open-source projects.

DeepSeek-R1-0528

MIT

Open weights

Step3-VL-10B

Apache 2.0

Open weights

Release Timeline

When each model was launched

DeepSeek-R1-0528 was released on 2025-05-28, while Step3-VL-10B was released on 2026-01-15.

Step3-VL-10B is 8 months newer than DeepSeek-R1-0528.

DeepSeek-R1-0528

May 28, 2025

1.3 years ago

Step3-VL-10B

Jan 15, 2026

8 months ago

7mo newer

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against DeepSeek-R1-0528 and Step3-VL-10B side-by-side, then vote on the output you prefer.

DeepSeek-R1-0528
✓ Preferred
Step3-VL-10B
Open in Playground

FAQ

Common questions about DeepSeek-R1-0528 vs Step3-VL-10B.

Which is better, DeepSeek-R1-0528 or Step3-VL-10B?

DeepSeek-R1-0528 and Step3-VL-10B are closely matched on the LLM Stats Score at 24.1 and 25.5. DeepSeek-R1-0528 is made by DeepSeek and Step3-VL-10B is made by StepFun. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does DeepSeek-R1-0528 compare to Step3-VL-10B in benchmarks?

DeepSeek-R1-0528 scores MMLU-Redux: 93.4%, SimpleQA: 92.3%, AIME 2024: 91.4%, AIME 2025: 87.5%, MMLU-Pro: 85.0%. Step3-VL-10B scores MMBench: 91.8%, AIME 2025: 87.7%, MathVista: 84.0%, MMMU: 78.1%, MathVision: 70.8%.

What are the context window sizes for DeepSeek-R1-0528 and Step3-VL-10B?

DeepSeek-R1-0528 supports 164K tokens and Step3-VL-10B supports an unknown number of tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between DeepSeek-R1-0528 and Step3-VL-10B?

Key differences include LLM Stats Score (24.1 vs 25.5), multimodal support (no vs yes), licensing (MIT vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes DeepSeek-R1-0528 and Step3-VL-10B?

DeepSeek-R1-0528 is developed by DeepSeek and Step3-VL-10B is developed by StepFun.