Navigate
Advertise here

DeepSeek V4 Flash 0731 (high) vs Gemini 3.5 Flash (high)

DeepSeek V4 Flash 0731 (high) leads on average score with 8.4 vs 8.3. DeepSeek V4 Flash 0731 (high) has the lower benchmark cost at $0.580 vs $2.737. Gemini 3.5 Flash (high) is faster at 18.59s vs 118.83s, with pass rates of 76.8% vs 85.5%.

Last updated at: 2026-10-02

Compared models

Rank
#74
Total Output Tokens
532,511
Response Time (avg)
118.83s
Total Cost
$0.580
Rank
#77
Total Output Tokens
246,012
Response Time (avg)
18.59s
Total Cost
$2.737
Recommended model DeepSeek V4 Flash 0731 (high)

It has the best score here (8.4), while costing about 4.7x less than Gemini 3.5 Flash (high).

Detailed comparison

Metric DeepSeek V4 Flash 0731 DeepSeek V4 Flash 0731 high Release: 2026-08-01 Gemini 3.5 Flash Gemini 3.5 Flash high Release: 2026-05-19
Score 8.4 8.3
Rank #74 #77
Reliability 10.0 10.0
Consistency 8.1 9.3
Attempts 69/69 69/69
Tests Correct
Attempt pass rate 76.8% 85.5%
Flaky tests 5 2
Total Runs 69 69
Cost per result 1.069 14.402
Total Cost $0.580 $2.737
Input Price $0.018 / 1M $1.500 / 1M
Output Price $1.280 / 1M $9.000 / 1M
Cache Read Price $0.018 / 1M $0.150 / 1M
Cache Write Price N/A $0.084 / 1M
Total Input Tokens 281,863 348,103
Output Tokens 24,535 11,045
Reasoning Tokens 507,976 234,967
Response Time (avg) 118.83s 18.59s
Response Time (max) 600.61s 145.92s
Response Time (total) 2733.01s 427.52s
Parameters 284B total (13B active) ~500B total (~20B active)
Availability Closed Closed

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.

#74 DeepSeek V4 Flash 0731

high
No output was saved. The original provider response or failure reason is unavailable.
Cost
$0.000
Time
187.3s
Tokens
19,048 tok

#77 Gemini 3.5 Flash

high
Cost
$0.208
Time
118.2s
Tokens
23,158 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 Flash 0731 6.3 6.3 55.6% 1 252.67s 7,044 2,160 191,210
Gemini 3.5 Flash 10.0 10.0 100.0% 0 22.96s 8,118 456 47,129

Quick Compare

Switch Comparison Pair