Head-to-head AI model comparison graphic: DeepSeek V4 Pro 0813 (DeepSeek) vs Step 5 Preview (StepFun). Intelligence index 23.1 vs 28.5; output speed 67 vs 151 tokens per second; time to first token 1.7s vs 0.91s; output price $3.96 vs $2.70 per million tokens; input price $1.32 vs $1.00 per million tokens. franklineh.com
DeepSeek V4 Pro 0813 vs Step 5 Preview: head-to-head AI benchmark comparison. Step 5 Preview leads on intelligence index, output speed, time to first token, output price and input price.
Head to headUpdated

DeepSeek V4 Pro 0813 Max vs Step 5 Preview compared

Step 5 Preview leads the intelligence index 28.5 to 23.1. Step 5 Preview costs 28% less on blended price. Benchmarks, sample bills, speed and a verdict.

Pick it for DeepSeek V4 Pro 0813 · Max

  • Teams already building on DeepSeek, if Step 5 Preview's recorded advantages are small for your workload.

Pick it for Step 5 Preview

  • Higher intelligence index: 28.5 against 23.1.
  • Faster output: 151 tokens per second against 67 tokens per second.
  • Lower cost: $1.54 against $2.11 for 1 million input and 200,000 output tokens.

By the numbers

MetricDeepSeek V4 Pro 0813 (Max)Step 5 Preview
Intelligence Index23.128.5
Input price (per 1M tokens)$1.32$1.00
Output price (per 1M tokens)$3.96$2.70
Output speed (tokens/s)67151
Time to first token1.7s0.91s
Release date2026-08-132026-09-18

DeepSeek V4 Pro 0813 (Max) (DeepSeek) and Step 5 Preview (StepFun) meet here on their recorded benchmark, price and speed data. What follows is built entirely from tracked data, so where the catalogue is silent, this page says so instead of guessing. Once text starts flowing, Step 5 Preview is the quicker writer by a wide margin, at 2.3 times the output speed of DeepSeek V4 Pro 0813 Max.

Step 5 Preview is the faster writer

Once a response is under way, DeepSeek V4 Pro 0813 Max produces roughly 67 tokens per second against 151 tokens per second for Step 5 Preview. On long outputs, that is a real advantage for Step 5 Preview. Time to first token is 1.7 seconds for DeepSeek V4 Pro 0813 Max and 0.91 seconds for Step 5 Preview. Step 5 Preview gets to its first token 1.9 times as fast.

Add the two numbers up for a 2,000-token reply and Step 5 Preview finishes first: 32 seconds for DeepSeek V4 Pro 0813 Max, 14 seconds for Step 5 Preview.

How the intelligence index compares

Step 5 Preview comes out ahead on the intelligence index, 28.5 to 23.1. That is a clear margin for an index that averages many evaluations. Put another way, DeepSeek V4 Pro 0813 Max would need 23% more on this index to draw level.

For context, that places DeepSeek V4 Pro 0813 Max 72nd and Step 5 Preview 61st out of 109 scored models still on offer.

The catalogue does not yet split either model's score into coding and math indexes.

What they cost

Per million tokens, DeepSeek V4 Pro 0813 Max charges $1.32 for input and $3.96 for output, while Step 5 Preview charges $1.00 and $2.70. That makes Step 5 Preview's output 32% cheaper.

For a sample workload of 1 million input tokens and 200,000 output tokens, DeepSeek V4 Pro 0813 Max would cost $2.11 and Step 5 Preview $1.54, a difference of $0.57. For a support assistant answering 10,000 conversations of 2,000 input and 500 output tokens each, Step 5 Preview would come in 27% lower: $46.20 for DeepSeek V4 Pro 0813 Max against $33.50 for Step 5 Preview.

When they came out

The release dates are August 13, 2026 for DeepSeek V4 Pro 0813 Max and September 18, 2026 for Step 5 Preview, 36 days apart. They are effectively contemporaries.

Value for money

Per point of intelligence index, DeepSeek V4 Pro 0813 Max costs $0.086 of blended price per million tokens and Step 5 Preview $0.050. Step 5 Preview scores higher and costs less per point, so on price per point there is nothing to trade off.

Quick answers

Is Step 5 Preview better than DeepSeek V4 Pro 0813 Max?

Where the data separates them, Step 5 Preview is ahead each time: Step 5 Preview scores higher on the intelligence index, 28.5 to 23.1; Step 5 Preview is 28% cheaper on blended price; and Step 5 Preview streams output 2.3 times as fast.

Does DeepSeek V4 Pro 0813 Max respond faster than Step 5 Preview?

Step 5 Preview generates output faster, at 151 tokens per second against 67 tokens per second. Step 5 Preview reaches its first token sooner (0.91 seconds against 1.7 seconds).

Where do these numbers come from?

The figures are the recorded benchmark, price and speed data in Franklin AI's model catalogue. This page is written from that data automatically; neither model was tested by hand for it.

Who makes DeepSeek V4 Pro 0813 Max and Step 5 Preview?

DeepSeek V4 Pro 0813 Max is made by DeepSeek and Step 5 Preview by StepFun. DeepSeek V4 Pro 0813 Max was released on August 13, 2026 and Step 5 Preview on September 18, 2026.

When were DeepSeek V4 Pro 0813 Max and Step 5 Preview released?

Step 5 Preview is newer. It was released on September 18, 2026, 36 days after DeepSeek V4 Pro 0813 Max (August 13, 2026).

Final whistle

The verdict: Step 5 Preview comes out ahead on every recorded measure where they differ, so it is the easier choice unless something outside this data matters to you. Pick Step 5 Preview if you need the higher overall score (28.5 against 23.1) and long outputs need to finish quickly. Pick DeepSeek V4 Pro 0813 Max only if your own tests show an advantage the recorded data does not.

Where the numbers come from

The scores, speed and prices in this comparison come from independent third-party model evaluations that Franklin AI syncs each day, and the article was written from a snapshot of them. For the latest results, open the DeepSeek V4 Pro 0813 (Max) and Step 5 Preview pages. The AI Benchmark methodology explains how that data is collected and shown.

Comments (0)

No comments yet

Be the first to share your thoughts!