Thoughts on AI, technology, and the future we're building.

New posts every week

HomeAll PostsAI NewsAI Basics
Timelines
ChatGPTOpenAI release historyAnthropic ClaudeClaude release historyGoogle GeminiGemini release history
Benchmarks
OverviewFull model trackerValue RankingsPerformance for the moneyCoding RankingsSWE-bench and code signalsAgent RankingsTool and workflow signalsReasoning RankingsKnowledge and reasoningLong ContextDocument and retrieval signalsLab ComparisonsProvider-level rankings
CategoriesAboutContact

Subscribe to Newsletter

Practical AI news, tips, tricks, tool analysis, sent straight to your inbox.

No spam. Unsubscribe anytime.

Practical explainers, tool notes, and systems thinking for people turning new AI capability into useful work.

Explore

  • All Posts
  • Categories
  • About
  • Contact

Categories

  • AI News
  • AI Basics
  • ChatGPT
  • Anthropic
  • AI Tools
  • AI Video
  • AI Images
  • Courses

Connect

LinkedInTwitterRSS

© 2026. All rights reserved.

Benchmark suiteUpdated May 27, 2026

Gemma 4 26b A4B.

A model-level benchmark profile for Gemma 4 26b A4B, combining Arena category ranks, suite indexes, commercial price and speed data, SWE-bench status, and lab context.

Google lab page

Lab

Google

Open weights

Coverage

67%

6 of 9 tracked signals

Best Arena rank

#31

Vision Arena

Best suite

#34

Value / 65

OverviewFull public trackerValuePerformance for the moneyCodingCode and SWE-bench signalsAgentsTool and workflow readinessReasoningKnowledge and reasoning signalsContextDocument and retrieval signalsLabsProvider comparisons

Model profile

Benchmark signal map.

This view treats missing values as coverage gaps, then puts every available score, rank, cost, and speed signal into the same visual hierarchy.

Analyst view

Gemma 4 26b A4B is strongest on value.

67% coverage

Its best composite result is rank #34 in the value view, with a 65 value index. The strongest raw Arena signal is Vision Arena at rank #31 with 10,707 public votes.

Best current lane

Value

#34 / 65 value index

Strongest raw signal

Vision

#31, 1241 score

Commercial posture

$0.198

0.77s first token / 62 tok/s

Coverage gaps

Search ArenaDocument ArenaSWE-bench

Profile completeness

6 of 9 signals matched

67%

Arena categories

3/5

Vision, Code, Text

Commercial match

Live

Gemma 4 26B A4B (Non-reasoning)

SWE-bench

Not listed

No confident software benchmark row

Raw identifier

gemma-4-26b-a4b

Open weights

Performance cockpit

Public benchmark scores

Arena rows show score, public rank, confidence interval, vote count, and publish date. Missing rows stay visible as coverage gaps.

Text Arena

Arena category

1,439#56

1431-1446 CI

5,789 votes

May 27, 2026

Arena public leaderboard dataset

WebDev Arena

Arena category

1,360#52

1344-1376 CI

1,502 votes

May 25, 2026

Arena public leaderboard dataset

Vision Arena

Arena category

1,241#31

1233-1249 CI

10,707 votes

May 18, 2026

Arena public leaderboard dataset

Search Arena

Arena category

Not listed

No confident public Arena row for this category.

Coverage gap

Document Arena

Arena category

Not listed

No confident public Arena row for this category.

Coverage gap

SWE-bench Verified

Software issue resolution

Not listed

No confident SWE-bench row matched to this model.

Coverage gap

Cost and speed

Commercial profile

Price, latency, and throughput are only shown when a confident commercial source row is matched.

Blended price

$0.198

USD / 1M tokens

Input price

$0.130

USD / 1M input tokens

Output price

$0.400

USD / 1M output tokens

Latency

0.77s

Median time to first token

Output speed

62 tok/s

Median generation speed

Matched source row

Gemma 4 26B A4B (Non-reasoning)

Suite position

Composite ranks by benchmark lens.

These views combine available public signals into purpose-specific indexes. A missing card means the model lacks enough source coverage for that lens.

Value

Quality, price, latency, and output speed

#3465

value index

blended / 1M tokens$0.198
time to first token0.77s
output speed62 tok/s

Coding

Code Arena, SWE-bench, and output speed

#6147

coding index

Code Arena rank #521,360
No confident matchNot listed
output speed62 tok/s

Agents

Search, documents, code, and latency

#7375

agent proxy

Search ArenaNot listed
Docs ArenaNot listed
latency0.77s

Reasoning

Text, documents, vision, and mode signal

#6145

reasoning proxy

Text Arena rank #561,439
Docs ArenaNot listed
Vision Arena rank #311,241

Long context

Document, search, and text benchmark proxy

#7631

context proxy

Docs ArenaNot listed
Search ArenaNot listed
Text Arena rank #561,439

Lab context

Nearby Google profiles.

Same-lab models make it easier to see whether this profile is a flagship, value play, specialist, or lower-coverage row.

Full lab

Gemini 3.1 Pro Grounding

gemini-3.1-pro-grounding

Proprietary

Best rank

#4

Coverage

11%

Gemini 3.1 Pro Preview

gemini-3.1-pro-preview

Proprietary

Best rank

#6

Coverage

67%

Gemini 3 Pro

gemini-3-pro

Proprietary

Best rank

#6

Coverage

56%

Gemini 3 Pro Grounding

gemini-3-pro-grounding

Proprietary

Best rank

#7

Coverage

11%

Gemini 3.5 Flash

gemini-3.5-flash

Proprietary

Best rank

#10

Coverage

56%

Gemini 3 Flash Grounding

gemini-3-flash-grounding

Proprietary

Best rank

#10

Coverage

11%

Gemini 3 Flash

gemini-3-flash (thinking-minimal)

Proprietary

Best rank

#17

Coverage

78%

Gemma 4 31b

gemma-4-31b

Open weights

Best rank

#23

Coverage

56%