Enter opens the highlighted result. Esc closes.

Claude Opus 4.8 vs DeepSeek V4 Flash 0731

Claude Opus 4.8 leads DeepSeek V4 Flash 0731 on coding, 65.0 to 51.8 on the Vibe Coding Index, and DeepSeek V4 Flash 0731 costs 89.3x less per output token ($0.28 vs $25.00 per million). Updated October 21, 2026; 51 benchmarks have a result for both.

Replace Claude Opus 4.8 Replace DeepSeek V4 Flash 0731
Claude Opus 4.8Anthropic
Coding
65.0
Overall
69.8
Output $/M
$25.00
47
DeepSeek V4 Flash 0731DeepSeek
Coding
51.8
Overall
55.2
Output $/M
$0.28
2
vs boards won of 51 shared, 2 tied

Claude Opus 4.8 or DeepSeek V4 Flash 0731: which to pick

Pick Claude Opus 4.8 for coding: it scores 65.0 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731.

Pick DeepSeek V4 Flash 0731 for output tokens 89.3x cheaper ($0.28 vs $25.00 per million) and a lower cost per solved task ($0.01 against $2.17).

Model AClaude Opus 4.8Anthropic65.0/1006 independent benchmarks$2.17 per solved taskModel BDeepSeek V4 Flash 0731DeepSeek51.8/1004 independent benchmarks$0.01 per solved task
Shared results
51
Claude Opus 4.8 only
79
DeepSeek V4 Flash 0731 only
4
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.

At a glance

Coding index
Claude Opus 4.8 65.0 from 6 benchmarks, DeepSeek V4 Flash 0731 51.8 from 4 benchmarks.
Widest category
Math, 17.8 points apart on average over 4 shared benchmarks.
Widest single gap
FrontierCode: Claude Opus 4.8 46.5% against DeepSeek V4 Flash 0731 18.8%, 42.6 points apart on a 0 to 100 scale.
Output price
Claude Opus 4.8 $25.00 against DeepSeek V4 Flash 0731 $0.28 per million output tokens, 89.3x apart.
Context window
Claude Opus 4.8 takes 1M tokens, DeepSeek V4 Flash 0731 1.05M.

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
CategoryClaude Opus 4.8BarsDeepSeek V4 Flash 0731Ties
Coding111
Intelligence20
Agentic work501 tie
Reasoning801 tie
Math40
Knowledge60
Web design80
General31
All shared boards4722 ties
CodingLike-for-likeSix of these boards feed the Vibe Coding Index111
Coding benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
LiveBench coding77.3%Not measuredOnly Claude Opus 4.8 measured
LMArena WebDev1556Not measuredOnly Claude Opus 4.8 measured
AA Coding Index74.369.1Claude Opus 4.8
LiveCodeBench (Vals AI)87.8%87.3%Claude Opus 4.8
SWE-bench Verified (Vals AI)88.6%88.8%DeepSeek V4 Flash 0731
Terminal-Bench 4 (Vals AI)23.2%18.7%Claude Opus 4.8
Vibe Code Bench (Vals AI)82.7%74.7%Claude Opus 4.8
Code migration (Vals AI)47.3%38.6%Claude Opus 4.8
ProgramBench (Vals AI)1.0%0.0%Claude Opus 4.8
IOI (Vals AI)Not measured32.7%Only DeepSeek V4 Flash 0731 measured
Terminal-Bench 2.1 (Vals AI)71.9%67.0%Claude Opus 4.8
SkillsBench (Vals AI)59.2%50.7%Claude Opus 4.8
CyberBench (Vals AI)Not measured72.1%Only DeepSeek V4 Flash 0731 measured
SciCode54.4%50.3%Claude Opus 4.8
Terminal-Bench Hard (Artificial Analysis run)58.3%Not measuredOnly Claude Opus 4.8 measured
FrontierCode46.5%18.8%Claude Opus 4.8
GSO-Bench47.1%Not measuredOnly Claude Opus 4.8 measured
WeirdML82.9%63.0%Claude Opus 4.8
ALE-BenchNot measured1306.1Only DeepSeek V4 Flash 0731 measured
LMArena Text: coding1490Not measuredOnly Claude Opus 4.8 measured
LMArena Text: software and IT1488Not measuredOnly Claude Opus 4.8 measured
LMArena WebDev: React1557Not measuredOnly Claude Opus 4.8 measured
LMArena WebDev: HTML1543Not measuredOnly Claude Opus 4.8 measured
SciCode12.3%Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 15 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
AA Intelligence Index41.834.3Claude Opus 4.8
Epoch Capabilities Index159.0154.8Claude Opus 4.8
Agentic workLike-for-likePublic boards, shown as published, kept out of the index50
Agentic work benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
AA Agentic Index41.941.0Claude Opus 4.8
Design Arena: agents building web apps1233Not measuredOnly Claude Opus 4.8 measured
Design Arena: agents building full-stack apps1249Not measuredOnly Claude Opus 4.8 measured
Design Arena: agents building mobile apps1222Not measuredOnly Claude Opus 4.8 measured
Design Arena: agents building Android apps1232Not measuredOnly Claude Opus 4.8 measured
Design Arena: agents building Godot games1231Not measuredOnly Claude Opus 4.8 measured
Design Arena: agents building games1238Not measuredOnly Claude Opus 4.8 measured
Design Arena: HTML slides1221Not measuredOnly Claude Opus 4.8 measured
Design Arena: PowerPoint slides1227Not measuredOnly Claude Opus 4.8 measured
Design Arena: spreadsheets1270Not measuredOnly Claude Opus 4.8 measured
LiveBench research81.8%Not measuredOnly Claude Opus 4.8 measured
LiveBench browser96.0%Not measuredOnly Claude Opus 4.8 measured
LiveBench long agentic48.3%Not measuredOnly Claude Opus 4.8 measured
Finance Agent (Vals AI)53.9%49.5%Claude Opus 4.8
Tax Agent Bench (Vals AI)30.0%28.1%Claude Opus 4.8
Legal Research Bench (Vals AI)43.8%30.3%Claude Opus 4.8
Excel Modeling Benchmark (Vals AI)69.4%57.0%Claude Opus 4.8
BioMysteryBench (Vals AI)Not measured64.4%Only DeepSeek V4 Flash 0731 measured
Terminal-Bench-Science (Vals AI)4.3%4.3%Tie
Time Horizon Index (Vals AI)11.8%Not measuredOnly Claude Opus 4.8 measured
τ²-Bench Telecom94.4%Not measuredOnly Claude Opus 4.8 measured
APEX-Agents48.9%Not measuredOnly Claude Opus 4.8 measured
Remote Labor Index8.3%Not measuredOnly Claude Opus 4.8 measured
OSWorld 2.020.6%Not measuredOnly Claude Opus 4.8 measured
DeepResearch Bench50.2%Not measuredOnly Claude Opus 4.8 measured
PostTrainBench33.8%Not measuredOnly Claude Opus 4.8 measured
Vending-Bench 2$5,787Not measuredOnly Claude Opus 4.8 measured
LMArena Search1204Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 13 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index80
Reasoning benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
GPQA Diamond92.4%89.9%Claude Opus 4.8
LiveBench reasoning89.2%Not measuredOnly Claude Opus 4.8 measured
MMMU Pro (Vals AI)86.6%Not measuredOnly Claude Opus 4.8 measured
GPQA Diamond (Artificial Analysis run)92.0%90.8%Claude Opus 4.8
Humanity's Last Exam (Artificial Analysis run)48.7%38.6%Claude Opus 4.8
GPQA Diamond (Epoch AI run)91.0%91.0%Tie
ARC-AGI92.5%89.0%Claude Opus 4.8
ARC-AGI-272.1%61.4%Claude Opus 4.8
SimpleBench64.8%61.1%Claude Opus 4.8
Chess puzzles52.0%32.0%Claude Opus 4.8
Mystery game puzzles36.0%34.0%Claude Opus 4.8
EBR-bench28.6%Not measuredOnly Claude Opus 4.8 measured
LMArena Text: hard prompts1482Not measuredOnly Claude Opus 4.8 measured
LMArena Text: expert1503Not measuredOnly Claude Opus 4.8 measured
LMArena Text: hard prompts in English1487Not measuredOnly Claude Opus 4.8 measured
LMArena Vision: diagrams1315Not measuredOnly Claude Opus 4.8 measured
LMArena Vision: homework1337Not measuredOnly Claude Opus 4.8 measured
Kagi LLM Benchmark88.8%Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index40
Math benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
OTIS Mock AIME98.3%94.4%Claude Opus 4.8
FrontierMath, tiers 1 to 380.0%57.5%Claude Opus 4.8
FrontierMath, tier 456.1%24.4%Claude Opus 4.8
ProofBench69.0%56.0%Claude Opus 4.8
LMArena Text: math1486Not measuredOnly Claude Opus 4.8 measured
LMArena Text: mathematical work1497Not measuredOnly Claude Opus 4.8 measured
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 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
MMLU Pro89.6%86.2%Claude Opus 4.8
LegalBench (Vals AI)83.6%77.7%Claude Opus 4.8
CorpFin (Vals AI)66.7%61.9%Claude Opus 4.8
TaxEval (Vals AI)75.6%70.7%Claude Opus 4.8
MortgageTax (Vals AI)69.9%Not measuredOnly Claude Opus 4.8 measured
MedCode (Vals AI)53.2%41.4%Claude Opus 4.8
SimpleQA Verified53.0%33.6%Claude Opus 4.8
LMArena Text: science1477Not measuredOnly Claude Opus 4.8 measured
LMArena Text: medicine and healthcare1464Not measuredOnly Claude Opus 4.8 measured
LMArena Text: legal and government1477Not measuredOnly Claude Opus 4.8 measured
LMArena Text: business and finance1458Not measuredOnly Claude Opus 4.8 measured
FACTS Benchmark Suite53.5%Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
Design Arena: websites12591238Claude Opus 4.8
Design Arena: UI components12571235Claude Opus 4.8
Design Arena: game dev12641208Claude Opus 4.8
Design Arena: data viz12401180Claude Opus 4.8
Design Arena: 3D12231209Claude Opus 4.8
Design Arena: all code categories12551230Claude Opus 4.8
Design Arena: SVG11941185Claude Opus 4.8
Design Arena: ASCII art12771090Claude Opus 4.8
LiveBench design78.8%Not measuredOnly Claude Opus 4.8 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index31
General benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
LiveBench data analysis66.0%Not measuredOnly Claude Opus 4.8 measured
Public Benefits Bench (Vals AI)68.1%Not measuredOnly Claude Opus 4.8 measured
SAGE (Vals AI)54.8%Not measuredOnly Claude Opus 4.8 measured
MedScribe (Vals AI)85.8%80.4%Claude Opus 4.8
Vals Multimodal Index70.9%Not measuredOnly Claude Opus 4.8 measured
IFBench62.2%Not measuredOnly Claude Opus 4.8 measured
Long context reasoning (AA-LCR)77.7%79.7%DeepSeek V4 Flash 0731
DTBench94.9%90.9%Claude Opus 4.8
LMCA57.5%41.7%Claude Opus 4.8
Surface Evolver Bench87.5%Not measuredOnly Claude Opus 4.8 measured
Furniture assembly42.5%Not measuredOnly Claude Opus 4.8 measured
LMArena Text1461Not measuredOnly Claude Opus 4.8 measured
LMArena Text: instruction following1476Not measuredOnly Claude Opus 4.8 measured
LMArena Text: creative writing1454Not measuredOnly Claude Opus 4.8 measured
LMArena Vision1294Not measuredOnly Claude Opus 4.8 measured
LMArena Text: English1468Not measuredOnly Claude Opus 4.8 measured
LMArena Text: non-English1450Not measuredOnly Claude Opus 4.8 measured
LMArena Text: Chinese1506Not measuredOnly Claude Opus 4.8 measured
LMArena Text: German1472Not measuredOnly Claude Opus 4.8 measured
LMArena Text: Japanese1442Not measuredOnly Claude Opus 4.8 measured
LMArena Text: Korean1431Not measuredOnly Claude Opus 4.8 measured
LMArena Text: French1481Not measuredOnly Claude Opus 4.8 measured
LMArena Text: Spanish1465Not measuredOnly Claude Opus 4.8 measured
LMArena Text: Russian1474Not measuredOnly Claude Opus 4.8 measured
LMArena Text: Polish1459Not measuredOnly Claude Opus 4.8 measured
LMArena Text: multi-turn1476Not measuredOnly Claude Opus 4.8 measured
LMArena Text: longer queries1483Not measuredOnly Claude Opus 4.8 measured
LMArena Text: writing and language1461Not measuredOnly Claude Opus 4.8 measured
LMArena Text: entertainment and media1440Not measuredOnly Claude Opus 4.8 measured
LMArena Vision: English1293Not measuredOnly Claude Opus 4.8 measured
LMArena Vision: Chinese1346Not measuredOnly Claude Opus 4.8 measured
LMArena Vision: reading text1303Not measuredOnly Claude Opus 4.8 measured
LMArena Vision: humor1282Not measuredOnly Claude Opus 4.8 measured
LMArena Vision: creative writing1301Not measuredOnly Claude Opus 4.8 measured
LMArena Document1475Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Opus 4.8 and DeepSeek V4 Flash 0731
BenchmarkClaude Opus 4.8DeepSeek V4 Flash 0731Leader
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 4.8$0.018Fits in one request
DeepSeek V4 Flash 0731$0.0003Fits in one request

DeepSeek V4 Flash 0731 costs less

Repository review

50K input + 3K output tokens

Claude Opus 4.8$0.325Fits in one request
DeepSeek V4 Flash 0731$0.0078Fits in one request

DeepSeek V4 Flash 0731 costs less

Agent loop

200K input + 10K output tokens

Claude Opus 4.8$1.25Fits in one request
DeepSeek V4 Flash 0731$0.031Fits in one request

DeepSeek V4 Flash 0731 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 4.8 and DeepSeek V4 Flash 0731
SpecificationAnthropic Claude Opus 4.8DeepSeek V4 Flash 0731
MakerAnthropicDeepSeek
WeightsNo public weights listedOpen weights
ReleasedMay 27, 2026July 31, 2026
Context window1M1.05M
Input $/Mtok$5.00$0.14
Output $/Mtok$25.00$0.28
Coding index65.051.8
Independent measurements behind it64
Cost per solved task$2.17$0.01
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 4.8 better than DeepSeek V4 Flash 0731 for coding?

Pick Claude Opus 4.8 for coding: it scores 65.0 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731. Pick DeepSeek V4 Flash 0731 for output tokens 89.3x cheaper ($0.28 vs $25.00 per million) and a lower cost per solved task ($0.01 against $2.17).

Which is cheaper, Claude Opus 4.8 or DeepSeek V4 Flash 0731?

Claude Opus 4.8 costs $5.00 per million input tokens and $25.00 per million output tokens; DeepSeek V4 Flash 0731 costs $0.14 and $0.28. Per solved SWE-bench Verified task, Claude Opus 4.8 costs $2.17 and DeepSeek V4 Flash 0731 $0.01.

Which has the bigger context window, Claude Opus 4.8 or DeepSeek V4 Flash 0731?

DeepSeek V4 Flash 0731, with 1.05M tokens against 1M.

How many benchmarks is the Claude Opus 4.8 vs DeepSeek V4 Flash 0731 comparison based on?

51 benchmarks have a result for both models as of October 21, 2026. Claude Opus 4.8's coding index rests on 6 independent benchmarks and DeepSeek V4 Flash 0731's on 4. 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-4-8-vs-deepseek-v4-flash-0731" title="Claude Opus 4.8 vs DeepSeek V4 Flash 0731, 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-4-8-vs-deepseek-v4-flash-0731">LLMBenchmark.ai, Claude Opus 4.8 vs DeepSeek V4 Flash 0731</a></p>