Enter opens the highlighted result. Esc closes.

Claude Fable 5 vs Claude Haiku 4.5

Claude Fable 5 leads Claude Haiku 4.5 on coding, 79.3 to 35.6 on the Vibe Coding Index, and Claude Haiku 4.5 costs 10.0x less per output token ($5.00 vs $50.00 per million). Updated October 22, 2026; 83 benchmarks have a result for both.

Replace Claude Fable 5 Replace Claude Haiku 4.5
Claude Fable 5Anthropic
Coding
79.3
Overall
79.0
Output $/M
$50.00
83
Claude Haiku 4.5Anthropic
Coding
35.6
Overall
41.0
Output $/M
$5.00
0
vs boards won of 83 shared

Claude Fable 5 or Claude Haiku 4.5: which to pick

Pick Claude Fable 5 for coding: it scores 79.3 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5.

Pick Claude Haiku 4.5 for output tokens 10.0x cheaper ($5.00 vs $50.00 per million) and a lower cost per solved task ($0.56 against $2.16).

Model AClaude Fable 5Anthropic79.3/1006 independent benchmarks$2.16 per solved taskModel BClaude Haiku 4.5Anthropic35.6/1004 independent benchmarks$0.56 per solved task
Shared results
83
Claude Fable 5 only
44
Claude Haiku 4.5 only
12
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 79.3 from 6 benchmarks, Claude Haiku 4.5 35.6 from 4 benchmarks.
Widest category
Agentic work, 40.2 points apart on average over 9 shared benchmarks.
Widest single gap
ARC-AGI-2: Claude Fable 5 89.2% against Claude Haiku 4.5 4.0%, 85.1 points apart on a 0 to 100 scale.
Output price
Claude Fable 5 $50.00 against Claude Haiku 4.5 $5.00 per million output tokens, 10.0x apart.
Context window
Claude Fable 5 takes 1M tokens, Claude Haiku 4.5 200K.

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 5BarsClaude Haiku 4.5Ties
Coding140
Intelligence20
Agentic work90
Reasoning100
Math30
Knowledge110
Web design90
General250
All shared boards830
CodingLike-for-likeSix of these boards feed the Vibe Coding Index140
Coding benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
SWE-bench VerifiedNot measured66.6%Only Claude Haiku 4.5 measured
SWE-bench MultilingualNot measured64.7%Only Claude Haiku 4.5 measured
LiveBench coding98.3%48.1%Claude Fable 5
LMArena WebDev16241330Claude Fable 5
AA Coding Index76.543.9Claude Fable 5
LiveCodeBench (Vals AI)89.8%41.2%Claude Fable 5
SWE-bench Verified (Vals AI)95.0%66.6%Claude Fable 5
Terminal-Bench 4 (Vals AI)41.4%Not measuredOnly Claude Fable 5 measured
Vibe Code Bench (Vals AI)90.3%11.4%Claude Fable 5
Code migration (Vals AI)55.1%7.5%Claude Fable 5
ProgramBench (Vals AI)2.0%0.0%Claude Fable 5
Terminal-Bench 2.1 (Vals AI)80.5%43.8%Claude Fable 5
SciCode61.0%Not measuredOnly Claude Fable 5 measured
Terminal-Bench Hard (Artificial Analysis run)62.9%Not measuredOnly Claude Fable 5 measured
Terminal-BenchNot measured35.5%Only Claude Haiku 4.5 measured
FrontierCode53.5%Not measuredOnly Claude Fable 5 measured
FrontierSWE47.0%Not measuredOnly Claude Fable 5 measured
GSO-Bench78.4%Not measuredOnly Claude Fable 5 measured
WeirdML91.9%45.4%Claude Fable 5
MirrorCode63.9%Not measuredOnly Claude Fable 5 measured
ALE-Bench2041.3Not measuredOnly Claude Fable 5 measured
LMArena Text: coding15181454Claude Fable 5
LMArena Text: software and IT15131439Claude Fable 5
LMArena WebDev: React16311326Claude Fable 5
LMArena WebDev: HTML16511318Claude Fable 5
LMArena WebDev: from an image1623Not measuredOnly Claude Fable 5 measured
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 Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
AA Intelligence Index49.616.9Claude Fable 5
Epoch Capabilities Index163.4142.2Claude Fable 5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index90
Agentic work benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
AA Agentic Index50.78.0Claude Fable 5
Design Arena: agents building web apps1261Not measuredOnly Claude Fable 5 measured
Design Arena: agents building full-stack apps1280Not measuredOnly Claude Fable 5 measured
Design Arena: agents building mobile apps1264Not measuredOnly Claude Fable 5 measured
Design Arena: agents building Android apps1284Not measuredOnly Claude Fable 5 measured
Design Arena: agents building Godot games1292Not measuredOnly Claude Fable 5 measured
Design Arena: agents building games1285Not measuredOnly Claude Fable 5 measured
Design Arena: HTML slides1227Not measuredOnly Claude Fable 5 measured
Design Arena: PowerPoint slides1270Not measuredOnly Claude Fable 5 measured
LiveBench research86.9%42.2%Claude Fable 5
LiveBench browser92.0%40.0%Claude Fable 5
LiveBench long agentic53.2%11.1%Claude Fable 5
Finance Agent (Vals AI)56.3%31.0%Claude Fable 5
Tax Agent Bench (Vals AI)38.3%5.9%Claude Fable 5
Legal Research Bench (Vals AI)49.5%10.6%Claude Fable 5
Excel Modeling Benchmark (Vals AI)73.7%22.7%Claude Fable 5
Terminal-Bench-Science (Vals AI)15.7%Not measuredOnly Claude Fable 5 measured
τ²-Bench Telecom98.5%Not measuredOnly Claude Fable 5 measured
APEX-Agents63.6%Not measuredOnly Claude Fable 5 measured
Remote Labor Index16.1%Not measuredOnly Claude Fable 5 measured
BALROGNot measured31.2%Only Claude Haiku 4.5 measured
DeepResearch BenchNot measured45.5%Only Claude Haiku 4.5 measured
PostTrainBench41.8%Not measuredOnly Claude Fable 5 measured
Vending-Bench 2$5,680$459Claude Fable 5
LMArena Search1229Not measuredOnly Claude Fable 5 measured
Berkeley Function Calling Leaderboard (BFCL)Not measured68.7%Only Claude Haiku 4.5 measured
Neither model has a result on the other 15 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index100
Reasoning benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
GPQA Diamond93.2%72.2%Claude Fable 5
LiveBench reasoning89.7%61.7%Claude Fable 5
MMMU Pro (Vals AI)89.3%46.1%Claude Fable 5
GPQA Diamond (Artificial Analysis run)92.6%Not measuredOnly Claude Fable 5 measured
Humanity's Last Exam (Artificial Analysis run)55.5%Not measuredOnly Claude Fable 5 measured
GPQA Diamond (Epoch AI run)85.9%71.2%Claude Fable 5
ARC-AGI98.5%47.7%Claude Fable 5
ARC-AGI-289.2%4.0%Claude Fable 5
SimpleBench81.9%Not measuredOnly Claude Fable 5 measured
Chess puzzles47.0%8.0%Claude Fable 5
Mystery game puzzles52.0%Not measuredOnly Claude Fable 5 measured
EBR-bench39.5%Not measuredOnly Claude Fable 5 measured
LMArena Text: hard prompts15071421Claude Fable 5
LMArena Text: expert15341443Claude Fable 5
LMArena Text: hard prompts in English15091435Claude Fable 5
LMArena Vision: diagrams1357Not measuredOnly Claude Fable 5 measured
LMArena Vision: homework1361Not measuredOnly Claude Fable 5 measured
Kagi LLM Benchmark91.4%Not measuredOnly Claude Fable 5 measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index30
Math benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
AIMENot measured82.7%Only Claude Haiku 4.5 measured
MGSM (Vals AI)Not measured92.2%Only Claude Haiku 4.5 measured
OTIS Mock AIME100.0%66.7%Claude Fable 5
FrontierMath, tiers 1 to 387.0%Not measuredOnly Claude Fable 5 measured
FrontierMath, tier 490.2%Not measuredOnly Claude Fable 5 measured
ProofBench95.0%Not measuredOnly Claude Fable 5 measured
MATH level 5Not measured96.4%Only Claude Haiku 4.5 measured
LMArena Text: math15161398Claude Fable 5
LMArena Text: mathematical work15151430Claude Fable 5
Neither model has a result on the other 4 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index110
Knowledge benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
MMLU Pro91.5%78.7%Claude Fable 5
LegalBench (Vals AI)88.6%81.2%Claude Fable 5
CaseLaw (Vals AI)Not measured56.5%Only Claude Haiku 4.5 measured
CorpFin (Vals AI)71.8%60.6%Claude Fable 5
TaxEval (Vals AI)76.9%67.5%Claude Fable 5
MortgageTax (Vals AI)68.9%62.2%Claude Fable 5
MedQA (Vals AI)Not measured79.6%Only Claude Haiku 4.5 measured
MedCode (Vals AI)56.1%32.7%Claude Fable 5
SimpleQA Verified70.7%13.2%Claude Fable 5
LMArena Text: science15061406Claude Fable 5
LMArena Text: medicine and healthcare14801394Claude Fable 5
LMArena Text: legal and government14981400Claude Fable 5
LMArena Text: business and finance14801400Claude Fable 5
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index90
Web design benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
Design Arena: websites12951122Claude Fable 5
Design Arena: UI components13161097Claude Fable 5
Design Arena: game dev13351097Claude Fable 5
Design Arena: data viz13041124Claude Fable 5
Design Arena: 3D13181071Claude Fable 5
Design Arena: all code categories13101114Claude Fable 5
Design Arena: SVG13021025Claude Fable 5
Design Arena: ASCII art13291140Claude Fable 5
LiveBench design84.7%59.9%Claude Fable 5
GeneralLike-for-likePublic boards, shown as published, kept out of the index250
General benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
LiveBench data analysis80.5%59.3%Claude Fable 5
Public Benefits Bench (Vals AI)70.4%54.3%Claude Fable 5
SAGE (Vals AI)51.9%31.8%Claude Fable 5
MedScribe (Vals AI)88.5%85.2%Claude Fable 5
Vals Multimodal Index74.2%42.9%Claude Fable 5
IFBench63.5%Not measuredOnly Claude Fable 5 measured
Long context reasoning (AA-LCR)82.3%Not measuredOnly Claude Fable 5 measured
DTBench98.4%73.6%Claude Fable 5
LMCA61.2%30.9%Claude Fable 5
Surface Evolver Bench95.0%Not measuredOnly Claude Fable 5 measured
Furniture assembly35.8%Not measuredOnly Claude Fable 5 measured
ExploitBenchNot measured13.7%Only Claude Haiku 4.5 measured
LMArena Text14911397Claude Fable 5
LMArena Text: instruction following15021414Claude Fable 5
LMArena Text: creative writing14941371Claude Fable 5
LMArena Vision1325Not measuredOnly Claude Fable 5 measured
LMArena Text: English14981414Claude Fable 5
LMArena Text: non-English14821377Claude Fable 5
LMArena Text: Chinese15431418Claude Fable 5
LMArena Text: German14861374Claude Fable 5
LMArena Text: Japanese15041338Claude Fable 5
LMArena Text: Korean14871346Claude Fable 5
LMArena Text: French15051411Claude Fable 5
LMArena Text: Spanish14981421Claude Fable 5
LMArena Text: Russian15031381Claude Fable 5
LMArena Text: Polish14941343Claude Fable 5
LMArena Text: multi-turn15041409Claude Fable 5
LMArena Text: longer queries15101427Claude Fable 5
LMArena Text: writing and language15011386Claude Fable 5
LMArena Text: entertainment and media14821369Claude Fable 5
LMArena Vision: English1323Not measuredOnly Claude Fable 5 measured
LMArena Vision: Chinese1373Not measuredOnly Claude Fable 5 measured
LMArena Vision: reading text1337Not measuredOnly Claude Fable 5 measured
LMArena Vision: humor1341Not measuredOnly Claude Fable 5 measured
LMArena Vision: creative writing1338Not measuredOnly Claude Fable 5 measured
LMArena Document14961420Claude Fable 5
Neither model has a result on the other 10 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Fable 5 and Claude Haiku 4.5
BenchmarkClaude Fable 5Claude Haiku 4.5Leader
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$0.035Fits in one request
Claude Haiku 4.5$0.0035Fits in one request

Claude Haiku 4.5 costs less

Repository review

50K input + 3K output tokens

Claude Fable 5$0.650Fits in one request
Claude Haiku 4.5$0.065Fits in one request

Claude Haiku 4.5 costs less

Agent loop

200K input + 10K output tokens

Claude Fable 5$2.50Fits in one request
Claude Haiku 4.5$0.250Exceeds the context window

Claude Haiku 4.5 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 and Claude Haiku 4.5
SpecificationAnthropic Claude Fable 5Anthropic Claude Haiku 4.5
MakerAnthropicAnthropic
WeightsNo public weights listedNo public weights listed
ReleasedJune 9, 2026October 15, 2025
Context window1M200K
Input $/Mtok$10.00$1.00
Output $/Mtok$50.00$5.00
Coding index79.335.6
Independent measurements behind it64
Cost per solved task$2.16$0.56
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Fable 5 better than Claude Haiku 4.5 for coding?

Pick Claude Fable 5 for coding: it scores 79.3 on the Vibe Coding Index against 35.6 for Claude Haiku 4.5. Pick Claude Haiku 4.5 for output tokens 10.0x cheaper ($5.00 vs $50.00 per million) and a lower cost per solved task ($0.56 against $2.16).

Which is cheaper, Claude Fable 5 or Claude Haiku 4.5?

Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Claude Haiku 4.5 costs $1.00 and $5.00. Per solved SWE-bench Verified task, Claude Fable 5 costs $2.16 and Claude Haiku 4.5 $0.56.

Which has the bigger context window, Claude Fable 5 or Claude Haiku 4.5?

Claude Fable 5, with 1M tokens against 200K.

How many benchmarks is the Claude Fable 5 vs Claude Haiku 4.5 comparison based on?

83 benchmarks have a result for both models as of October 22, 2026. Claude Fable 5's coding index rests on 6 independent benchmarks and Claude Haiku 4.5'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-fable-5-vs-claude-haiku-4-5" title="Claude Fable 5 vs Claude Haiku 4.5, 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-vs-claude-haiku-4-5">LLMBenchmark.ai, Claude Fable 5 vs Claude Haiku 4.5</a></p>