Claude Opus 5 vs Gemma 4 31B

Claude Opus 5 leads Gemma 4 31B on coding, 83.0 to 39.0 on the Vibe Coding Index, and Gemma 4 31B costs 62.5x less per output token ($0.40 vs $25.00 per million). Updated October 22, 2026; 62 benchmarks have a result for both.

Replace Claude Opus 5 Replace Gemma 4 31B

Claude Opus 5Anthropic

Coding: 83.0
Overall: 81.6
Output $/M: $25.00

Gemma 4 31BGoogle

Coding: 39.0
Overall: 42.1
Output $/M: $0.40

vs boards won of 62 shared

Claude Opus 5 or Gemma 4 31B: which to pick

Pick Claude Opus 5 for coding: it scores 83.0 on the Vibe Coding Index against 39.0 for Gemma 4 31B.

Pick Gemma 4 31B for output tokens 62.5x cheaper ($0.40 vs $25.00 per million) and work measured by SAGE (Vals AI), where it leads.

Model AClaude Opus 5Anthropic83.0/1006 independent benchmarks$1.33 per solved task Model BGemma 4 31BGoogle39.0/1002 independent benchmarksNo cost per solved task

62699

Shared results: 62
Claude Opus 5 only: 69
Gemma 4 31B only: 9
Like-for-like categories: 8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

Coding
Vibe Coding Index, six public coding benchmarks
Claude Opus 5
Claude Opus 5 scores 83.0 against 39.0.
6 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
Claude Opus 5
Claude Opus 5 leads, 50.8 against 14.7.
Same board, both measured
Agentic work
AA Agentic Index
Claude Opus 5
Claude Opus 5 leads, 56.5 against 4.2.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Claude Opus 5 has a GPQA Diamond result, 93.4%.
Directional only
Knowledge
MMLU Pro
No call
Only Claude Opus 5 has a MMLU Pro result, 91.6%.
Directional only
Web design
Design Arena: websites
No call
Only Claude Opus 5 has a Design Arena: websites result, 1309.
Directional only
Long documents
Listed context window
Claude Opus 5
Claude Opus 5 takes 1M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Gemma 4 31B
Claude Opus 5 $0.018, Gemma 4 31B $0.0003 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Claude Opus 5 has the number, $1.33.
Directional only

At a glance

Coding index: Claude Opus 5 83.0 from 6 benchmarks, Gemma 4 31B 39.0 from 2 benchmarks.
Widest category: Agentic work, 56.4 points apart on average over 4 shared benchmarks.
Widest single gap: LiveBench browser: Claude Opus 5 84.0% against Gemma 4 31B 4.0%, 80.0 points apart on a 0 to 100 scale.
Output price: Claude Opus 5 $25.00 against Gemma 4 31B $0.40 per million output tokens, 62.5x apart.
Context window: Claude Opus 5 takes 1M tokens, Gemma 4 31B 262K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
Category	Claude Opus 5	Gemma 4 31B
Coding	11	0
Intelligence	2	0
Agentic work	4	0
Reasoning	9	0
Math	3	0
Knowledge	6	0
Web design	1	0
General	25	1
All shared boards	61	1

CodingLike-for-likeSix of these boards feed the Vibe Coding Index110

Coding benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
LiveBench coding	90.8%	49.0%	Claude Opus 5
LMArena WebDev	1689	1367	Claude Opus 5
AA Coding Index	78.0	43.4	Claude Opus 5
LiveCodeBench (Vals AI)	89.0%	Not measured	Only Claude Opus 5 measured
SWE-bench Verified (Vals AI)	97.0%	Not measured	Only Claude Opus 5 measured
Terminal-Bench 4 (Vals AI)	53.5%	Not measured	Only Claude Opus 5 measured
Vibe Code Bench (Vals AI)	88.4%	Not measured	Only Claude Opus 5 measured
Code migration (Vals AI)	57.5%	Not measured	Only Claude Opus 5 measured
ProgramBench (Vals AI)	3.0%	Not measured	Only Claude Opus 5 measured
IOI (Vals AI)	84.3%	Not measured	Only Claude Opus 5 measured
Terminal-Bench 2.1 (Vals AI)	84.6%	Not measured	Only Claude Opus 5 measured
Vibe Code Bench 1-100 (Vals AI)	28.5%	Not measured	Only Claude Opus 5 measured
SkillsBench (Vals AI)	60.4%	Not measured	Only Claude Opus 5 measured
CyberBench (Vals AI)	65.4%	Not measured	Only Claude Opus 5 measured
SciCode	56.4%	45.5%	Claude Opus 5
Terminal-Bench Hard (Artificial Analysis run)	Not measured	36.4%	Only Gemma 4 31B measured
FrontierCode	53.4%	Not measured	Only Claude Opus 5 measured
FrontierSWE	52.0%	Not measured	Only Claude Opus 5 measured
WeirdML	91.8%	52.3%	Claude Opus 5
ALE-Bench	2164.6	925.5	Claude Opus 5
LMArena Text: coding	1533	1459	Claude Opus 5
LMArena Text: software and IT	1527	1463	Claude Opus 5
LMArena WebDev: React	1711	1359	Claude Opus 5
LMArena WebDev: HTML	1680	1380	Claude Opus 5
LMArena WebDev: from an image	1662	Not measured	Only Claude Opus 5 measured
SciCode	13.8%	8.9%	Claude Opus 5
Neither model has a result on the other 13 boards here

IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20

Intelligence benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
AA Intelligence Index	50.8	14.7	Claude Opus 5
Epoch Capabilities Index	163.9	143.0	Claude Opus 5

Agentic workLike-for-likePublic boards, shown as published, kept out of the index40

Agentic work benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
AA Agentic Index	56.5	4.2	Claude Opus 5
Design Arena: agents building web apps	1266	Not measured	Only Claude Opus 5 measured
Design Arena: agents building full-stack apps	1306	Not measured	Only Claude Opus 5 measured
Design Arena: agents building mobile apps	1348	Not measured	Only Claude Opus 5 measured
Design Arena: agents building Android apps	1257	Not measured	Only Claude Opus 5 measured
Design Arena: agents building Godot games	1279	Not measured	Only Claude Opus 5 measured
Design Arena: agents building games	1261	Not measured	Only Claude Opus 5 measured
Design Arena: PowerPoint slides	1240	Not measured	Only Claude Opus 5 measured
LiveBench research	88.9%	44.4%	Claude Opus 5
LiveBench browser	84.0%	4.0%	Claude Opus 5
LiveBench long agentic	62.1%	13.3%	Claude Opus 5
SREBench (Vals AI)	12.2%	Not measured	Only Claude Opus 5 measured
Finance Agent (Vals AI)	58.6%	Not measured	Only Claude Opus 5 measured
Tax Agent Bench (Vals AI)	45.5%	Not measured	Only Claude Opus 5 measured
Legal Research Bench (Vals AI)	55.3%	Not measured	Only Claude Opus 5 measured
Excel Modeling Benchmark (Vals AI)	73.6%	Not measured	Only Claude Opus 5 measured
BioMysteryBench (Vals AI)	79.3%	Not measured	Only Claude Opus 5 measured
Terminal-Bench-Science (Vals AI)	27.1%	Not measured	Only Claude Opus 5 measured
Vals Science Index	47.9%	Not measured	Only Claude Opus 5 measured
Time Horizon Index (Vals AI)	18.8%	Not measured	Only Claude Opus 5 measured
CUA-bench (Vals AI)	9.0%	Not measured	Only Claude Opus 5 measured
τ²-Bench Telecom	Not measured	65.5%	Only Gemma 4 31B measured
APEX-Agents	65.8%	Not measured	Only Claude Opus 5 measured
OSWorld 2.0	31.4%	Not measured	Only Claude Opus 5 measured
BALROG	63.4%	Not measured	Only Claude Opus 5 measured
PostTrainBench	35.0%	Not measured	Only Claude Opus 5 measured
Vending-Bench 2	$11,182	Not measured	Only Claude Opus 5 measured
LMArena Search	1242	Not measured	Only Claude Opus 5 measured
Neither model has a result on the other 13 boards here

ReasoningLike-for-likePublic boards, shown as published, kept out of the index90

Reasoning benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
GPQA Diamond	93.4%	Not measured	Only Claude Opus 5 measured
LiveBench reasoning	91.2%	68.3%	Claude Opus 5
MysteryMechanism (Vals AI)	37.4%	Not measured	Only Claude Opus 5 measured
MMMU Pro (Vals AI)	89.9%	Not measured	Only Claude Opus 5 measured
GPQA Diamond (Artificial Analysis run)	93.7%	85.7%	Claude Opus 5
Humanity's Last Exam (Artificial Analysis run)	54.9%	23.6%	Claude Opus 5
GPQA Diamond (Epoch AI run)	93.9%	75.8%	Claude Opus 5
ARC-AGI	97.5%	Not measured	Only Claude Opus 5 measured
ARC-AGI-2	90.4%	Not measured	Only Claude Opus 5 measured
SimpleBench	80.6%	Not measured	Only Claude Opus 5 measured
Chess puzzles	43.0%	Not measured	Only Claude Opus 5 measured
Mystery game puzzles	59.0%	Not measured	Only Claude Opus 5 measured
EBR-bench	45.7%	Not measured	Only Claude Opus 5 measured
LMArena Text: hard prompts	1527	1448	Claude Opus 5
LMArena Text: expert	1554	1464	Claude Opus 5
LMArena Text: hard prompts in English	1523	1456	Claude Opus 5
LMArena Vision: diagrams	1348	1283	Claude Opus 5
LMArena Vision: homework	1338	1310	Claude Opus 5
Kagi LLM Benchmark	Not measured	63.5%	Only Gemma 4 31B measured
Neither model has a result on the other 3 boards here

MathLike-for-likePublic boards, shown as published, kept out of the index30

Math benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
OTIS Mock AIME	98.9%	73.3%	Claude Opus 5
FrontierMath, tiers 1 to 3	85.6%	Not measured	Only Claude Opus 5 measured
FrontierMath, tier 4	73.2%	Not measured	Only Claude Opus 5 measured
ProofBench	99.0%	Not measured	Only Claude Opus 5 measured
LMArena Text: math	1533	1463	Claude Opus 5
LMArena Text: mathematical work	1544	1465	Claude Opus 5
Neither model has a result on the other 7 boards here

KnowledgeLike-for-likePublic boards, shown as published, kept out of the index60

Knowledge benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
MMLU Pro	91.6%	Not measured	Only Claude Opus 5 measured
LegalBench (Vals AI)	87.0%	Not measured	Only Claude Opus 5 measured
CaseLaw (Vals AI)	Not measured	52.6%	Only Gemma 4 31B measured
CorpFin (Vals AI)	73.2%	Not measured	Only Claude Opus 5 measured
TaxEval (Vals AI)	75.1%	Not measured	Only Claude Opus 5 measured
MortgageTax (Vals AI)	72.1%	61.4%	Claude Opus 5
MedCode (Vals AI)	63.6%	Not measured	Only Claude Opus 5 measured
SimpleQA Verified	59.9%	10.4%	Claude Opus 5
LMArena Text: science	1531	1457	Claude Opus 5
LMArena Text: medicine and healthcare	1525	1451	Claude Opus 5
LMArena Text: legal and government	1530	1438	Claude Opus 5
LMArena Text: business and finance	1501	1430	Claude Opus 5
FACTS Benchmark Suite	Not measured	45.3%	Only Gemma 4 31B measured
Neither model has a result on the other 5 boards here

Web designLike-for-likePublic boards, shown as published, kept out of the index10

Web design benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
Design Arena: websites	1309	Not measured	Only Claude Opus 5 measured
Design Arena: UI components	1346	Not measured	Only Claude Opus 5 measured
Design Arena: game dev	1337	Not measured	Only Claude Opus 5 measured
Design Arena: data viz	1329	Not measured	Only Claude Opus 5 measured
Design Arena: 3D	1334	Not measured	Only Claude Opus 5 measured
Design Arena: all code categories	1323	Not measured	Only Claude Opus 5 measured
Design Arena: SVG	1311	Not measured	Only Claude Opus 5 measured
Design Arena: ASCII art	1352	Not measured	Only Claude Opus 5 measured
LiveBench design	83.0%	58.4%	Claude Opus 5

GeneralLike-for-likePublic boards, shown as published, kept out of the index251

General benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
LiveBench data analysis	74.6%	53.3%	Claude Opus 5
Public Benefits Bench (Vals AI)	76.9%	Not measured	Only Claude Opus 5 measured
SAGE (Vals AI)	49.4%	55.0%	Gemma 4 31B
MedScribe (Vals AI)	91.0%	Not measured	Only Claude Opus 5 measured
Vals Multimodal Index	73.9%	Not measured	Only Claude Opus 5 measured
IFBench	Not measured	75.6%	Only Gemma 4 31B measured
Long context reasoning (AA-LCR)	82.0%	69.7%	Claude Opus 5
DTBench	97.9%	82.7%	Claude Opus 5
LMCA	64.5%	39.3%	Claude Opus 5
Surface Evolver Bench	Not measured	30.6%	Only Gemma 4 31B measured
Furniture assembly	60.8%	Not measured	Only Claude Opus 5 measured
LMArena Text	1509	1443	Claude Opus 5
LMArena Text: instruction following	1518	1433	Claude Opus 5
LMArena Text: creative writing	1493	1417	Claude Opus 5
LMArena Vision	1320	1276	Claude Opus 5
LMArena Text: English	1508	1451	Claude Opus 5
LMArena Text: non-English	1502	1430	Claude Opus 5
LMArena Text: Chinese	1573	1475	Claude Opus 5
LMArena Text: German	1527	Not measured	Only Claude Opus 5 measured
LMArena Text: Japanese	1520	Not measured	Only Claude Opus 5 measured
LMArena Text: Korean	1517	Not measured	Only Claude Opus 5 measured
LMArena Text: French	1519	1436	Claude Opus 5
LMArena Text: Spanish	1523	1442	Claude Opus 5
LMArena Text: Russian	1507	1460	Claude Opus 5
LMArena Text: Polish	1522	1433	Claude Opus 5
LMArena Text: multi-turn	1500	1452	Claude Opus 5
LMArena Text: longer queries	1517	1446	Claude Opus 5
LMArena Text: writing and language	1499	1432	Claude Opus 5
LMArena Text: entertainment and media	1481	1405	Claude Opus 5
LMArena Vision: English	1315	1271	Claude Opus 5
LMArena Vision: Chinese	1400	1326	Claude Opus 5
LMArena Vision: reading text	1329	1282	Claude Opus 5
LMArena Vision: humor	1293	1280	Claude Opus 5
LMArena Vision: captioning	Not measured	1218	Only Gemma 4 31B measured
LMArena Vision: recognizing things	Not measured	1226	Only Gemma 4 31B measured
LMArena Vision: creative writing	1313	1285	Claude Opus 5
LMArena Document	1516	1425	Claude Opus 5
Neither model has a result on the other 9 boards here

Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared

Image generation benchmarks for Claude Opus 5 and Gemma 4 31B
Benchmark	Claude Opus 5	Gemma 4 31B	Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Claude Opus 5$0.018Fits in one request

Gemma 4 31B$0.0003Fits in one request

Gemma 4 31B costs less

Repository review

50K input + 3K output tokens

Claude Opus 5$0.325Fits in one request

Gemma 4 31B$0.0082Fits in one request

Gemma 4 31B costs less

Agent loop

200K input + 10K output tokens

Claude Opus 5$1.25Fits in one request

Gemma 4 31B$0.032Fits in one request

Gemma 4 31B costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Claude Opus 5 and Gemma 4 31B
Specification	Anthropic Claude Opus 5	Google Gemma 4 31B
Maker	Anthropic	Google
Weights	No public weights listed	Open weights
Released	July 24, 2026	April 2, 2026
Context window	1M	262K
Input $/Mtok	$5.00	$0.14
Output $/Mtok	$25.00	$0.40
Coding index	83.0	39.0
Independent measurements behind it	6	2
Cost per solved task	$1.33	Not computable
Source	OpenRouter listing	OpenRouter listing

Questions

Is Claude Opus 5 better than Gemma 4 31B for coding?

Pick Claude Opus 5 for coding: it scores 83.0 on the Vibe Coding Index against 39.0 for Gemma 4 31B. Pick Gemma 4 31B for output tokens 62.5x cheaper ($0.40 vs $25.00 per million) and work measured by SAGE (Vals AI), where it leads.

Which is cheaper, Claude Opus 5 or Gemma 4 31B?

Claude Opus 5 costs $5.00 per million input tokens and $25.00 per million output tokens; Gemma 4 31B costs $0.14 and $0.40. Cost per solved task is known for Claude Opus 5 only.

Which has the bigger context window, Claude Opus 5 or Gemma 4 31B?

Claude Opus 5, with 1M tokens against 262K.

How many benchmarks is the Claude Opus 5 vs Gemma 4 31B comparison based on?

62 benchmarks have a result for both models as of October 22, 2026. Claude Opus 5's coding index rests on 6 independent benchmarks and Gemma 4 31B's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/claude-opus-5-vs-gemma-4-31b-it" title="Claude Opus 5 vs Gemma 4 31B, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/claude-opus-5-vs-gemma-4-31b-it">LLMBenchmark.ai, Claude Opus 5 vs Gemma 4 31B</a></p>

Claude Opus 5 or Gemma 4 31B: which to pick

Which one for your work

Coding

Intelligence

Agentic work

Reasoning

Knowledge

Web design

Long documents

Everyday cost

Cost per solved task

At a glance

Benchmark by benchmark

What each workload costs

Chat turn

Repository review

Agent loop

Specification differences

Questions

Compare next