Navigate
AI BENCHY
Advertise here

DeepSeek V4 Pro (high) vs Step 3.7 Flash (low)

DeepSeek V4 Pro (high) leads on average score with 7.7 vs 7.1. Step 3.7 Flash (low) has the lower benchmark cost at $0.390 vs $0.761. Step 3.7 Flash (low) is faster at 17.87s vs 92.50s, with pass rates of 63.6% vs 63.6%.

Last updated at: 2026-08-20

Rank
#75
Total Output Tokens
209,444
Response Time (avg)
92.50s
Total Cost
$0.761
Rank
#113
Total Output Tokens
320,393
Response Time (avg)
17.87s
Total Cost
$0.390
Recommended model Step 3.7 Flash (low)

Its score stays close to the best score here (7.1 vs 7.7), while costing about 2.0x less than DeepSeek V4 Pro (high).

Detailed comparison

Metric DeepSeek V4 Pro DeepSeek V4 Pro high Release: 2026-04-24 Step 3.7 Flash Step 3.7 Flash low Release: 2026-05-29
Score 7.7 7.1
Rank #75 #113
Reliability 10.0 10.0
Consistency 7.7 8.1
Attempts 66/66 66/66
Tests Correct
Attempt pass rate 63.6% 63.6%
Flaky tests 6 5
Total Runs 66 66
Cost per result 2.251 3.539
Total Cost $0.761 $0.390
Input Price $1.600 / 1M $0.200 / 1M
Output Price $3.200 / 1M $1.150 / 1M
Total Input Tokens 90,757 103,842
Output Tokens 22,928 320,393
Reasoning Tokens 186,516 0
Response Time (avg) 92.50s 17.87s
Response Time (max) 416.76s 124.75s
Response Time (total) 2035.01s 393.04s
Parameters 1.6T total (49B active) 198B total (11B active)
Availability Open source Open source

Model generation showcase

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#75 DeepSeek V4 Pro

high
Cost
$0.023
Time
257.6s
Tokens
14,870 tok

#113 Step 3.7 Flash

low
Invalid SVG
Cost
$0.004
Time
25.3s
Tokens
3,072 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
DeepSeek V4 Pro 6.3 8.7 33.3% 0 243.00s 5,090 383 84,580
Step 3.7 Flash 8.2 7.2 88.9% 1 9.46s 7,437 18,685 0

Quick Compare

Switch Comparison Pair