DeepSeek V4 Pro 0424 vs Gemma 4 26B A4B: Release Comparison

Comparison of the DeepSeek V4 Pro 0424 and Gemma 4 26B A4B releases: 3 and 2 models respectively, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across all 5 models.

  • For intelligence, DeepSeek V4 Pro 0424 scores highest: DeepSeek V4 Pro (Reasoning, Max Effort) at 31, against Gemma 4 26B A4B (Reasoning) at 17 for Gemma 4 26B A4B.
  • For output speed, DeepSeek V4 Pro 0424 is fastest: DeepSeek V4 Pro (Non-reasoning) at 91 t/s, against Gemma 4 26B A4B (Non-reasoning) at 64 t/s for Gemma 4 26B A4B.

Releases compared

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3 incorporates 10 evaluations: AA-Briefcase, GDPval-AA v2, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1
Estimate (independent evaluation forthcoming)

Intelligence Index vs. Cost per Intelligence Index Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task
Most attractive quadrant
Pareto line

Cost

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better

Speed & Latency

Output Speed

Output tokens per second · Higher is better

Capability Scores

Capability Indices

Measures the performance of models on specific capabilities and industries
Finance & Accounting Index

Incorporates 7 evaluations: AA-Omniscience, GDPval-AA v2, AA-Briefcase, Humanity's Last Exam, AutomationBench-AA, AA-LCR v1.1, GDP.pdf · Higher is better

Strategy & Ops Index

Incorporates 6 evaluations: AA-Omniscience, GDPval-AA v2, AA-Briefcase, AutomationBench-AA, AA-LCR v1.1, GDP.pdf · Higher is better

Legal Index

Incorporates 7 evaluations: AA-Omniscience, GDPval-AA v2, AA-Briefcase, Humanity's Last Exam, AA-LCR v1.1, GDP.pdf, AutomationBench-AA · Higher is better

Healthcare & Medical Index

Incorporates 6 evaluations: AA-Omniscience, GDPval-AA v2, AA-Briefcase, MLCR-AA, Humanity's Last Exam, AutomationBench-AA · Higher is better

Engineering Index

Incorporates 6 evaluations: AA-Omniscience, Humanity's Last Exam, CritPt, GDPval-AA v2, AA-Briefcase, Terminal-Bench 4.0 · Higher is better

Economics Index

Incorporates 5 evaluations: AA-Omniscience, Humanity's Last Exam, GDPval-AA v2, AA-Briefcase, AA-LCR v1.1 · Higher is better

Further details

Input modality
Output modality
Weights
Provider Benchmarks
DeepSeek V4 Pro (Reasoning, Max Effort)
DeepSeek logoDeepSeek
31
35
40
34
33
32
38
$0.12
$0.2
$0.435
$0.87
$0.0036
$277
49k
33k
212M
86
1.61s
52.77s
58.61s
406.49s
1.6T
49B active at inference time
1M
Apr 2026
-
Yes

Supports: text

Supports: text

🤗
MIT
DeepInfraMicrosoft AzureNovita
+6
View
DeepSeek V4 Pro (Reasoning, High Effort)
DeepSeek logoDeepSeek
30
-
-
-
-
-
-
-
$0.2
$0.435
$0.87
$0.0036
-
-
-
-
90
1.71s
23.75s
29.28s
-
1.6T
49B active at inference time
1M
Apr 2026
-
Yes

Supports: text

Supports: text

🤗
MIT
DeepInfraMicrosoft AzureDeepSeek
+5
View
DeepSeek V4 Pro (Non-reasoning)
DeepSeek logoDeepSeek
21
-
-
-
-
-
-
-
$0.2
$0.435
$0.87
$0.0036
-
-
-
-
91
1.56s
1.56s
7.04s
-
1.6T
49B active at inference time
1M
Apr 2026
-
No

Supports: text

Supports: text

🤗
Mit
Microsoft AzureBitdeer AINebius
+2
View
Gemma 4 26B A4B (Reasoning)
Google logoGoogle
17
-
-
-
-
-
-
-
$0.1
$0.115
$0.37
$0.085
-
-
-
-
-
-
-
-
-
25.2B
3.8B active at inference time
256k
Apr 2026
-
Yes

Supports: text, image, and video

Supports: text

🤗
Apache 2.0
MakoraNovitaGoogle
+4
View
Gemma 4 26B A4B (Non-reasoning)
Google logoGoogle
13
-
-
-
-
-
-
-
$0.2
$0.13
$0.40
$0.125
-
-
-
-
64
1.46s
1.46s
9.22s
-
25.2B
3.8B active at inference time
256k
Apr 2026
-
No

Supports: text, image, and video

Supports: text

🤗
Apache 2.0
MakoraParasailGMI
+4
View