Navigate
Advertise here

Qwen3.8 2.4T A95B (high) vs Step 3.7 Flash (medium)

Qwen3.8 2.4T A95B (high) leads on average score with 7.9 vs 7.8. Step 3.7 Flash (medium) has the lower benchmark cost at $0.571 vs $2.949. Step 3.7 Flash (medium) is faster at 26.57s vs 120.88s, with pass rates of 75.4% vs 68.1%.

Last updated at: 2026-11-16

Compared models

Rank
#163
Total Output Tokens
368,836
Response Time (avg)
120.88s
Total Cost
$2.949
Rank
#170
Total Output Tokens
440,835
Response Time (avg)
26.57s
Total Cost
$0.571
Recommended model Step 3.7 Flash (medium)

Its score stays close to the best score here (7.8 vs 7.9), while costing about 5.2x less than Qwen3.8 2.4T A95B (high).

Detailed comparison

Metric Qwen3.8 2.4T A95B Qwen3.8 2.4T A95B high Release: 2026-08-12 Step 3.7 Flash Step 3.7 Flash medium Release: 2026-05-29
Score 7.9 7.8
Rank #163 #170
Reliability 9.6 10.0
Consistency 8.4 8.7
Attempts 69/69 69/69
Tests Correct
Attempt pass rate 75.4% 68.1%
Flaky tests 4 4
Total Runs 69 69
Cost per result 18.696 4.392
Total Cost $2.949 $0.571
Input Price $2.000 / 1M $0.200 / 1M
Output Price $6.000 / 1M $1.150 / 1M
Cache Read Price $0.250 / 1M $0.040 / 1M
Cache Write Price N/A N/A
Total Input Tokens 295,642 319,524
Output Tokens 115,041 440,835
Reasoning Tokens 353,398 0
Response Time (avg) 120.88s 26.57s
Response Time (max) 532.28s 152.83s
Response Time (total) 2780.16s 611.09s
Parameters 2.4T total (95B active) 198B total (11B active)
Availability Weights available Open source

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Model generation showcase

Hamster playing table tennis

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

#163 Qwen3.8 2.4T A95B

high
Reached the allocated time limit (600 seconds) without receiving showcase output.
Cost
$0.000
Time
600.0s
Tokens
0 tok

#170 Step 3.7 Flash

medium
Cost
$0.006
Time
46.2s
Tokens
4,466 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
Qwen3.8 2.4T A95B 5.9 6.3 55.6% 1 160.52s 6,399 3,310 48,078
Step 3.7 Flash 8.8 7.8 88.9% 1 27.42s 7,437 44,797 0

Switch Comparison Pair