Enter opens the highlighted result. Esc closes.

Claude Fable 5.1 vs DeepSeek V4 Pro 0423

Claude Fable 5.1 leads DeepSeek V4 Pro 0423 on coding, 84.2 to 38.7 on the Vibe Coding Index, and DeepSeek V4 Pro 0423 costs 18.4x less per output token ($2.716 vs $50.00 per million). Updated October 21, 2026; 77 benchmarks have a result for both.

Replace Claude Fable 5.1 Replace DeepSeek V4 Pro 0423
Claude Fable 5.1Anthropic
Coding
84.2
Overall
84.7
Output $/M
$50.00
77
DeepSeek V4 Pro 0423DeepSeek
Coding
38.7
Overall
50.0
Output $/M
$2.716
0
vs boards won of 77 shared

Claude Fable 5.1 or DeepSeek V4 Pro 0423: which to pick

Pick Claude Fable 5.1 for coding: it scores 84.2 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423.

Pick DeepSeek V4 Pro 0423 for output tokens 18.4x cheaper ($2.716 vs $50.00 per million) and a larger context window (1.05M against 1M tokens).

Model AClaude Fable 5.1Anthropic84.2/1005 independent benchmarksNo cost per solved taskModel BDeepSeek V4 Pro 0423DeepSeek38.7/1005 independent benchmarks$0.57 per solved task
Shared results
77
Claude Fable 5.1 only
47
DeepSeek V4 Pro 0423 only
7
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 Fable 5.1 84.2 from 5 benchmarks, DeepSeek V4 Pro 0423 38.7 from 5 benchmarks.
Widest category
Math, 41.1 points apart on average over 6 shared benchmarks.
Widest single gap
FrontierMath, tier 4: Claude Fable 5.1 87.8% against DeepSeek V4 Pro 0423 2.4%, 85.4 points apart on a 0 to 100 scale.
Output price
Claude Fable 5.1 $50.00 against DeepSeek V4 Pro 0423 $2.716 per million output tokens, 18.4x apart.
Context window
Claude Fable 5.1 takes 1M tokens, DeepSeek V4 Pro 0423 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 Fable 5.1BarsDeepSeek V4 Pro 0423Ties
Coding160
Intelligence20
Agentic work90
Reasoning60
Math60
Knowledge90
Web design80
General210
All shared boards770
CodingLike-for-likeSix of these boards feed the Vibe Coding Index160
Coding benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
LiveBench coding97.7%Not measuredOnly Claude Fable 5.1 measured
LMArena WebDev17431447Claude Fable 5.1
AA Coding Index81.659.4Claude Fable 5.1
LiveCodeBench (Vals AI)90.5%87.5%Claude Fable 5.1
SWE-bench Verified (Vals AI)Not measured77.4%Only DeepSeek V4 Pro 0423 measured
Terminal-Bench 4 (Vals AI)58.1%11.1%Claude Fable 5.1
Vibe Code Bench (Vals AI)90.3%49.9%Claude Fable 5.1
Code migration (Vals AI)54.6%26.2%Claude Fable 5.1
ProgramBench (Vals AI)7.0%0.0%Claude Fable 5.1
IOI (Vals AI)90.8%Not measuredOnly Claude Fable 5.1 measured
Terminal-Bench 2.1 (Vals AI)85.0%50.2%Claude Fable 5.1
SWE-bench Multimodal (Vals AI)61.3%Not measuredOnly Claude Fable 5.1 measured
Vibe Code Bench 1-100 (Vals AI)28.0%Not measuredOnly Claude Fable 5.1 measured
SkillsBench (Vals AI)61.5%51.3%Claude Fable 5.1
CyberBench (Vals AI)70.4%Not measuredOnly Claude Fable 5.1 measured
SciCode63.1%Not measuredOnly Claude Fable 5.1 measured
FrontierCode50.9%17.6%Claude Fable 5.1
FrontierSWE56.3%Not measuredOnly Claude Fable 5.1 measured
GSO-Bench88.2%Not measuredOnly Claude Fable 5.1 measured
WeirdML92.9%48.9%Claude Fable 5.1
MirrorCode73.3%Not measuredOnly Claude Fable 5.1 measured
ALE-Bench2143.21006.1Claude Fable 5.1
LMArena Text: coding15321470Claude Fable 5.1
LMArena Text: software and IT15301469Claude Fable 5.1
LMArena WebDev: React17921441Claude Fable 5.1
LMArena WebDev: HTML16841435Claude Fable 5.1
LMArena WebDev: from an image1722Not measuredOnly Claude Fable 5.1 measured
Neither model has a result on the other 12 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
AA Intelligence Index53.430.4Claude Fable 5.1
Epoch Capabilities Index166.5149.1Claude Fable 5.1
Agentic workLike-for-likePublic boards, shown as published, kept out of the index90
Agentic work benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
AA Agentic Index57.926.3Claude Fable 5.1
Design Arena: agents building web apps13181000Claude Fable 5.1
Design Arena: agents building full-stack apps1326948Claude Fable 5.1
Design Arena: agents building mobile apps1267Not measuredOnly Claude Fable 5.1 measured
Design Arena: agents building Godot games13271059Claude Fable 5.1
Design Arena: PowerPoint slides1327Not measuredOnly Claude Fable 5.1 measured
Design Arena: spreadsheets1365Not measuredOnly Claude Fable 5.1 measured
LiveBench research82.0%Not measuredOnly Claude Fable 5.1 measured
LiveBench browser100.0%Not measuredOnly Claude Fable 5.1 measured
LiveBench long agentic80.8%Not measuredOnly Claude Fable 5.1 measured
SREBench (Vals AI)22.9%Not measuredOnly Claude Fable 5.1 measured
Finance Agent (Vals AI)58.9%44.1%Claude Fable 5.1
Tax Agent Bench (Vals AI)49.2%25.2%Claude Fable 5.1
Legal Research Bench (Vals AI)55.3%23.1%Claude Fable 5.1
Excel Modeling Benchmark (Vals AI)76.7%51.6%Claude Fable 5.1
Terminal-Bench-Science (Vals AI)40.0%Not measuredOnly Claude Fable 5.1 measured
Time Horizon Index (Vals AI)63.3%Not measuredOnly Claude Fable 5.1 measured
CUA-bench (Vals AI)13.2%Not measuredOnly Claude Fable 5.1 measured
Robo-bench (Vals AI)5.8%Not measuredOnly Claude Fable 5.1 measured
APEX-Agents68.6%Not measuredOnly Claude Fable 5.1 measured
Remote Labor Index17.9%Not measuredOnly Claude Fable 5.1 measured
SpireClimb2.0%Not measuredOnly Claude Fable 5.1 measured
Vending-Bench 2$5,422$3,285Claude Fable 5.1
LMArena Search1246Not measuredOnly Claude Fable 5.1 measured
Neither model has a result on the other 17 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
GPQA Diamond93.4%89.4%Claude Fable 5.1
LiveBench reasoning91.7%Not measuredOnly Claude Fable 5.1 measured
MysteryMechanism (Vals AI)47.8%Not measuredOnly Claude Fable 5.1 measured
MMMU Pro (Vals AI)90.6%Not measuredOnly Claude Fable 5.1 measured
GPQA Diamond (Artificial Analysis run)93.7%Not measuredOnly Claude Fable 5.1 measured
Humanity's Last Exam (Artificial Analysis run)59.1%Not measuredOnly Claude Fable 5.1 measured
GPQA Diamond (Epoch AI run)Not measured90.9%Only DeepSeek V4 Pro 0423 measured
ARC-AGI97.5%Not measuredOnly Claude Fable 5.1 measured
ARC-AGI-290.0%Not measuredOnly Claude Fable 5.1 measured
Chess puzzles48.0%19.0%Claude Fable 5.1
Mystery game puzzles58.0%17.0%Claude Fable 5.1
EBR-bench57.1%Not measuredOnly Claude Fable 5.1 measured
LMArena Text: hard prompts15281461Claude Fable 5.1
LMArena Text: expert15441463Claude Fable 5.1
LMArena Text: hard prompts in English15221466Claude Fable 5.1
LMArena Vision: diagrams1343Not measuredOnly Claude Fable 5.1 measured
LMArena Vision: homework1345Not measuredOnly Claude Fable 5.1 measured
Kagi LLM BenchmarkNot measured53.5%Only DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index60
Math benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
OTIS Mock AIME100.0%96.7%Claude Fable 5.1
FrontierMath, tiers 1 to 390.2%45.3%Claude Fable 5.1
FrontierMath, tier 487.8%2.4%Claude Fable 5.1
ProofBench100.0%16.0%Claude Fable 5.1
LMArena Text: math15311443Claude Fable 5.1
LMArena Text: mathematical work15411455Claude Fable 5.1
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index90
Knowledge benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
MMLU Pro92.4%87.3%Claude Fable 5.1
LegalBench (Vals AI)88.5%80.3%Claude Fable 5.1
CaseLaw (Vals AI)Not measured59.4%Only DeepSeek V4 Pro 0423 measured
CorpFin (Vals AI)Not measured61.4%Only DeepSeek V4 Pro 0423 measured
TaxEval (Vals AI)76.0%72.1%Claude Fable 5.1
MortgageTax (Vals AI)70.8%Not measuredOnly Claude Fable 5.1 measured
MedCode (Vals AI)53.5%40.5%Claude Fable 5.1
SimpleQA Verified70.8%47.0%Claude Fable 5.1
LMArena Text: science15321470Claude Fable 5.1
LMArena Text: medicine and healthcare15311466Claude Fable 5.1
LMArena Text: legal and government15141464Claude Fable 5.1
LMArena Text: business and finance14931443Claude Fable 5.1
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 Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
Design Arena: websites13131234Claude Fable 5.1
Design Arena: UI components13391222Claude Fable 5.1
Design Arena: game dev13811229Claude Fable 5.1
Design Arena: data viz13441203Claude Fable 5.1
Design Arena: 3D13931254Claude Fable 5.1
Design Arena: all code categories13361240Claude Fable 5.1
Design Arena: SVG13011140Claude Fable 5.1
Design Arena: ASCII art13231149Claude Fable 5.1
LiveBench design83.4%Not measuredOnly Claude Fable 5.1 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index210
General benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
LiveBench data analysis80.3%Not measuredOnly Claude Fable 5.1 measured
Public Benefits Bench (Vals AI)74.9%62.9%Claude Fable 5.1
SAGE (Vals AI)48.5%Not measuredOnly Claude Fable 5.1 measured
MedScribe (Vals AI)91.3%75.1%Claude Fable 5.1
Long context reasoning (AA-LCR)85.3%Not measuredOnly Claude Fable 5.1 measured
DTBench97.6%90.7%Claude Fable 5.1
LMCA65.5%41.2%Claude Fable 5.1
Surface Evolver BenchNot measured40.0%Only DeepSeek V4 Pro 0423 measured
Furniture assembly70.0%Not measuredOnly Claude Fable 5.1 measured
CL-bench LifeNot measured13.5%Only DeepSeek V4 Pro 0423 measured
LMArena Text15121451Claude Fable 5.1
LMArena Text: instruction following15181445Claude Fable 5.1
LMArena Text: creative writing15091446Claude Fable 5.1
LMArena Vision1319Not measuredOnly Claude Fable 5.1 measured
LMArena Text: English15061459Claude Fable 5.1
LMArena Text: non-English15081439Claude Fable 5.1
LMArena Text: Chinese15841481Claude Fable 5.1
LMArena Text: German14991458Claude Fable 5.1
LMArena Text: Japanese15621446Claude Fable 5.1
LMArena Text: Korean15171409Claude Fable 5.1
LMArena Text: French15231473Claude Fable 5.1
LMArena Text: Spanish15201457Claude Fable 5.1
LMArena Text: Russian15201453Claude Fable 5.1
LMArena Text: Polish15251461Claude Fable 5.1
LMArena Text: multi-turn14931468Claude Fable 5.1
LMArena Text: longer queries15241458Claude Fable 5.1
LMArena Text: writing and language15101447Claude Fable 5.1
LMArena Text: entertainment and media14971431Claude Fable 5.1
LMArena Vision: English1313Not measuredOnly Claude Fable 5.1 measured
LMArena Vision: Chinese1391Not measuredOnly Claude Fable 5.1 measured
LMArena Vision: reading text1327Not measuredOnly Claude Fable 5.1 measured
LMArena Vision: humor1318Not measuredOnly Claude Fable 5.1 measured
LMArena Vision: creative writing1330Not measuredOnly Claude Fable 5.1 measured
Neither model has a result on the other 13 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Fable 5.1 and DeepSeek V4 Pro 0423
BenchmarkClaude Fable 5.1DeepSeek V4 Pro 0423Leader
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 Fable 5.1$0.035Fits in one request
DeepSeek V4 Pro 0423$0.0026Fits in one request

DeepSeek V4 Pro 0423 costs less

Repository review

50K input + 3K output tokens

Claude Fable 5.1$0.650Fits in one request
DeepSeek V4 Pro 0423$0.069Fits in one request

DeepSeek V4 Pro 0423 costs less

Agent loop

200K input + 10K output tokens

Claude Fable 5.1$2.50Fits in one request
DeepSeek V4 Pro 0423$0.272Fits in one request

DeepSeek V4 Pro 0423 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 Fable 5.1 and DeepSeek V4 Pro 0423
SpecificationAnthropic Claude Fable 5.1DeepSeek V4 Pro 0423
MakerAnthropicDeepSeek
WeightsNo public weights listedOpen weights
ReleasedSeptember 1, 2026April 24, 2026
Context window1M1.05M
Input $/Mtok$10.00$1.225
Output $/Mtok$50.00$2.716
Coding index84.238.7
Independent measurements behind it55
Cost per solved taskNot computable$0.57
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Fable 5.1 better than DeepSeek V4 Pro 0423 for coding?

Pick Claude Fable 5.1 for coding: it scores 84.2 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423. Pick DeepSeek V4 Pro 0423 for output tokens 18.4x cheaper ($2.716 vs $50.00 per million) and a larger context window (1.05M against 1M tokens).

Which is cheaper, Claude Fable 5.1 or DeepSeek V4 Pro 0423?

Claude Fable 5.1 costs $10.00 per million input tokens and $50.00 per million output tokens; DeepSeek V4 Pro 0423 costs $1.225 and $2.716. Cost per solved task is known for DeepSeek V4 Pro 0423 only.

Which has the bigger context window, Claude Fable 5.1 or DeepSeek V4 Pro 0423?

DeepSeek V4 Pro 0423, with 1.05M tokens against 1M.

How many benchmarks is the Claude Fable 5.1 vs DeepSeek V4 Pro 0423 comparison based on?

77 benchmarks have a result for both models as of October 21, 2026. Claude Fable 5.1's coding index rests on 5 independent benchmarks and DeepSeek V4 Pro 0423's on 5. 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-fable-5-1-vs-deepseek-v4-pro" title="Claude Fable 5.1 vs DeepSeek V4 Pro 0423, 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-fable-5-1-vs-deepseek-v4-pro">LLMBenchmark.ai, Claude Fable 5.1 vs DeepSeek V4 Pro 0423</a></p>