The AI arena is free today

Open Superagent

Atria Dawn Preview vs Step-3.5-Flash

Atria Dawn Preview leads the LLM Stats Score 52.3 to 37.2.

Shanghai AI Laboratory · StepFun · Updated for 2026

Which is better?

Atria Dawn Preview leads the overall LLM Stats Score 52.3 to 37.2, ranking #10 overall.

In the 1 individual benchmarks reported for both models, Atria Dawn Preview 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 Atria Dawn Preview

  • overall performance matters — it scores 52.3 and ranks #10 on LLM Stats
  • your work emphasizes reasoning and coding — it leads those capability indexes
  • you value its reported benchmark strengths — it wins 1 of 1 exact shared results
  • you want the most recent training data — it shipped Sep 2026

Choose Step-3.5-Flash

  • you want predictable pricing at $0.10/M input and $0.40/M output

At a glance

The differences that matter most.

Atria Dawn Preview
Shanghai AI Laboratory
Step-3.5-Flash
StepFun
Core performance indexes
LLM Stats Score
52.3
#10
37.2
#69
Reasoning
50.5
#12
37.2
#69
Coding
36.2
#22
20.5
#96
Agents
38.8
#10
15.3
#86
Cost, coverage & limits
Benchmark wins
1 of 1
0 of 1
Input price
— / M
$0.10 / M
Output price
— / M
$0.40 / M
Context window
—
65,536

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
Atria Dawn Preview
Step-3.5-Flash
Tool use
31.1#12
13.3#97
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

12 reported for Atria Dawn Preview · 7 for Step-3.5-Flash

1 shared

Atria Dawn Preview outperforms in 1 benchmarks (BrowseComp), while Step-3.5-Flash is better at 0 benchmarks.

Atria Dawn Preview significantly outperforms across most benchmarks.

Sun Sep 20 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

548.0B diff

Atria Dawn Preview has 548.0B more parameters than Step-3.5-Flash, making it 279.6% larger.

Shanghai AI Laboratory
Atria Dawn Preview
744.0Bparameters
StepFun
Step-3.5-Flash
196.0Bparameters
744.0B
Atria Dawn Preview
196.0B
Step-3.5-Flash

Context Window

Maximum input and output token capacity

Only Step-3.5-Flash specifies input context (65,536 tokens). Only Step-3.5-Flash specifies output context (8,192 tokens).

Shanghai AI Laboratory
Atria Dawn Preview
Input- tokens
Output- tokens
StepFun
Step-3.5-Flash
Input65,536 tokens
Output8,192 tokens
Sun Sep 20 2026 • llm-stats.com

License

Usage and distribution terms

Atria Dawn Preview is licensed under MIT, while Step-3.5-Flash uses Apache 2.0.

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

Atria Dawn Preview

MIT

Open weights

Step-3.5-Flash

Apache 2.0

Open weights

Release Timeline

When each model was launched

Atria Dawn Preview was released on 2026-09-11, while Step-3.5-Flash was released on 2026-02-02.

Atria Dawn Preview is 7 months newer than Step-3.5-Flash.

Atria Dawn Preview

Sep 11, 2026

1 weeks ago

7mo newer
Step-3.5-Flash

Feb 2, 2026

7 months ago

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 Atria Dawn Preview and Step-3.5-Flash side-by-side, then vote on the output you prefer.

Atria Dawn Preview
✓ Preferred
Step-3.5-Flash
Open in Playground

FAQ

Common questions about Atria Dawn Preview vs Step-3.5-Flash.

Which is better, Atria Dawn Preview or Step-3.5-Flash?

Atria Dawn Preview leads the LLM Stats Score 52.3 to 37.2. Atria Dawn Preview is made by Shanghai AI Laboratory and Step-3.5-Flash is made by StepFun. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Atria Dawn Preview compare to Step-3.5-Flash in benchmarks?

Atria Dawn Preview scores DeepSearchQA: 96.0%, BrowseComp: 92.5%, CyberGym: 86.5%, MLE-Bench Lite: 86.2%, WideSearch: 81.9%. Step-3.5-Flash scores AIME 2025: 97.3%, Tau-bench: 88.2%, LiveCodeBench v6: 86.4%, IMO-AnswerBench: 85.4%, SWE-Bench Verified: 74.4%.

What are the context window sizes for Atria Dawn Preview and Step-3.5-Flash?

Atria Dawn Preview supports an unknown number of tokens and Step-3.5-Flash supports 66K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Atria Dawn Preview and Step-3.5-Flash?

Key differences include LLM Stats Score (52.3 vs 37.2), licensing (MIT vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes Atria Dawn Preview and Step-3.5-Flash?

Atria Dawn Preview is developed by Shanghai AI Laboratory and Step-3.5-Flash is developed by StepFun.

Related comparisons