LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Claude Fable 5 vs Qwen3.8 27B

Updated October 21, 2026. 82 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Claude Fable 5 Replace Qwen3.8 27B
Claude Fable 5Anthropic
Coding
82.1
Overall
49.6
Output $/M
$50.00
81
Qwen3.8 27BQwen
Coding
73.6
Overall
33.7
Output $/M
$1.35
1
vs boards won of 82 shared

Decision reading

Pick Claude Fable 5 for coding: it scores 82.1 on the Vibe Coding Index against 73.6 for Qwen3.8 27B.

Pick Qwen3.8 27B when output tokens cost 37.0x less, or when a solved task costs $1.26 against $2.16.

Model AClaude Fable 5Anthropic82.1/1002 independent benchmarks$2.16 per solved taskModel BQwen3.8 27BQwen73.6/1002 independent benchmarks$1.26 per solved task
Shared results
82
Claude Fable 5 only
44
Qwen3.8 27B only
3
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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 82.1 from 2 benchmarks, Qwen3.8 27B 73.6 from 2 benchmarks.
Widest category
Math, 33.0 points apart on average over 3 shared benchmarks.
Widest single gap
ProofBench: Claude Fable 5 95.0% against Qwen3.8 27B 16.0%, 79.0 points apart on a 0 to 100 scale.
Output price
Claude Fable 5 $50.00 against Qwen3.8 27B $1.35 per million output tokens, 37.0x apart.
Context window
Both take 1M tokens.

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 5BarsQwen3.8 27BTies
Coding150
Intelligence20
Agentic work100
Reasoning120
Math30
Knowledge90
Web design10
General291
All shared boards811
CodingLike-for-likeFour of these boards feed the Vibe Coding Index150
Coding benchmarks for Claude Fable 5 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
LiveBench coding98.3%86.8%Claude Fable 5
LMArena WebDev16241593Claude Fable 5
AA Coding Index76.568.1Claude Fable 5
LiveCodeBench (Vals AI)89.8%84.0%Claude Fable 5
SWE-bench Verified (Vals AI)95.0%86.0%Claude Fable 5
Terminal-Bench 4 (Vals AI)41.4%Not measuredOnly Claude Fable 5 measured
Vibe Code Bench (Vals AI)90.3%64.8%Claude Fable 5
Code migration (Vals AI)55.1%14.2%Claude Fable 5
ProgramBench (Vals AI)2.0%0.0%Claude Fable 5
IOI (Vals AI)Not measured39.1%Only Qwen3.8 27B measured
Terminal-Bench 2.1 (Vals AI)80.5%58.4%Claude Fable 5
SkillsBench (Vals AI)Not measured38.1%Only Qwen3.8 27B measured
CyberBench (Vals AI)Not measured67.8%Only Qwen3.8 27B measured
SciCode61.0%46.6%Claude Fable 5
Terminal-Bench Hard (Artificial Analysis run)62.9%Not measuredOnly Claude Fable 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%Not measuredOnly Claude Fable 5 measured
MirrorCode63.9%Not measuredOnly Claude Fable 5 measured
ALE-Bench2041.3Not measuredOnly Claude Fable 5 measured
LMArena Text: coding15181483Claude Fable 5
LMArena Text: software and IT15131478Claude Fable 5
LMArena WebDev: React16311602Claude Fable 5
LMArena WebDev: HTML16511570Claude Fable 5
LMArena WebDev: from an image16231577Claude Fable 5
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 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
AA Intelligence Index49.633.7Claude Fable 5
Epoch Capabilities Index163.4151.7Claude Fable 5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index100
Agentic work benchmarks for Claude Fable 5 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
AA Agentic Index50.745.8Claude 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 apps1289Not 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 slides1277Not measuredOnly Claude Fable 5 measured
Design Arena: PowerPoint slides1295Not measuredOnly Claude Fable 5 measured
LiveBench research86.9%59.4%Claude Fable 5
LiveBench browser92.0%32.0%Claude Fable 5
LiveBench long agentic53.2%27.9%Claude Fable 5
Finance Agent (Vals AI)56.3%48.5%Claude Fable 5
Tax Agent Bench (Vals AI)38.3%30.9%Claude Fable 5
Legal Research Bench (Vals AI)49.5%36.1%Claude Fable 5
Excel Modeling Benchmark (Vals AI)73.7%59.7%Claude Fable 5
Terminal-Bench-Science (Vals AI)15.7%0.0%Claude Fable 5
τ²-Bench Telecom98.5%Not measuredOnly Claude Fable 5 measured
APEX-Agents63.6%47.5%Claude Fable 5
Remote Labor Index16.1%Not measuredOnly Claude Fable 5 measured
PostTrainBench41.8%Not measuredOnly Claude Fable 5 measured
Vending-Bench 2$5,680Not measuredOnly Claude Fable 5 measured
LMArena Search1229Not measuredOnly Claude Fable 5 measured
Neither model has a result on the other 16 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index120
Reasoning benchmarks for Claude Fable 5 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
GPQA Diamond93.2%88.9%Claude Fable 5
LiveBench reasoning89.7%80.0%Claude Fable 5
MMMU Pro (Vals AI)89.3%83.9%Claude Fable 5
GPQA Diamond (Artificial Analysis run)92.6%90.5%Claude Fable 5
Humanity's Last Exam (Artificial Analysis run)55.5%33.9%Claude Fable 5
GPQA Diamond (Epoch AI run)85.9%Not measuredOnly Claude Fable 5 measured
ARC-AGI98.5%87.5%Claude Fable 5
ARC-AGI-289.2%42.4%Claude Fable 5
SimpleBench81.9%Not measuredOnly Claude Fable 5 measured
Chess puzzles47.0%Not measuredOnly Claude Fable 5 measured
Mystery game puzzles52.0%Not measuredOnly Claude Fable 5 measured
EBR-bench39.5%Not measuredOnly Claude Fable 5 measured
LMArena Text: hard prompts15071459Claude Fable 5
LMArena Text: expert15341479Claude Fable 5
LMArena Text: hard prompts in English15091461Claude Fable 5
LMArena Vision: diagrams13571285Claude Fable 5
LMArena Vision: homework13611295Claude Fable 5
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 Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
OTIS Mock AIME100.0%Not measuredOnly Claude Fable 5 measured
FrontierMath, tiers 1 to 387.0%Not measuredOnly Claude Fable 5 measured
FrontierMath, tier 490.2%Not measuredOnly Claude Fable 5 measured
ProofBench95.0%16.0%Claude Fable 5
LMArena Text: math15161455Claude Fable 5
LMArena Text: mathematical work15151458Claude Fable 5
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 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
MMLU Pro91.5%84.3%Claude Fable 5
LegalBench (Vals AI)88.6%82.4%Claude Fable 5
CorpFin (Vals AI)71.8%Not measuredOnly Claude Fable 5 measured
TaxEval (Vals AI)76.9%70.8%Claude Fable 5
MortgageTax (Vals AI)68.9%64.9%Claude Fable 5
MedCode (Vals AI)56.1%28.7%Claude Fable 5
SimpleQA Verified70.7%Not measuredOnly Claude Fable 5 measured
LMArena Text: science15061467Claude Fable 5
LMArena Text: medicine and healthcare14801437Claude Fable 5
LMArena Text: legal and government14981447Claude Fable 5
LMArena Text: business and finance14801438Claude Fable 5
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index10
Web design benchmarks for Claude Fable 5 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
Design Arena: websites1296Not measuredOnly Claude Fable 5 measured
Design Arena: UI components1318Not measuredOnly Claude Fable 5 measured
Design Arena: game dev1336Not measuredOnly Claude Fable 5 measured
Design Arena: data viz1304Not measuredOnly Claude Fable 5 measured
Design Arena: 3D1319Not measuredOnly Claude Fable 5 measured
Design Arena: all code categories1310Not measuredOnly Claude Fable 5 measured
Design Arena: SVG1302Not measuredOnly Claude Fable 5 measured
Design Arena: ASCII art1328Not measuredOnly Claude Fable 5 measured
LiveBench design84.7%75.9%Claude Fable 5
GeneralLike-for-likePublic boards, shown as published, kept out of the index291
General benchmarks for Claude Fable 5 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
LiveBench data analysis80.5%76.6%Claude Fable 5
Public Benefits Bench (Vals AI)70.4%Not measuredOnly Claude Fable 5 measured
SAGE (Vals AI)51.9%52.4%Qwen3.8 27B
MedScribe (Vals AI)88.5%83.8%Claude Fable 5
Vals Multimodal Index74.2%Not measuredOnly Claude Fable 5 measured
IFBench63.5%Not measuredOnly Claude Fable 5 measured
Long context reasoning (AA-LCR)82.3%82.0%Claude Fable 5
DTBench98.4%88.0%Claude Fable 5
LMCA61.2%41.4%Claude Fable 5
Surface Evolver Bench95.0%45.0%Claude Fable 5
Furniture assembly35.8%Not measuredOnly Claude Fable 5 measured
LMArena Text14911440Claude Fable 5
LMArena Text: instruction following15021436Claude Fable 5
LMArena Text: creative writing14941383Claude Fable 5
LMArena Vision13251269Claude Fable 5
LMArena Text: English14981448Claude Fable 5
LMArena Text: non-English14821428Claude Fable 5
LMArena Text: Chinese15431501Claude Fable 5
LMArena Text: German14861443Claude Fable 5
LMArena Text: Japanese15041387Claude Fable 5
LMArena Text: Korean14871384Claude Fable 5
LMArena Text: French15051465Claude Fable 5
LMArena Text: Spanish14981449Claude Fable 5
LMArena Text: Russian15031415Claude Fable 5
LMArena Text: Polish14941439Claude Fable 5
LMArena Text: multi-turn15041440Claude Fable 5
LMArena Text: longer queries15101448Claude Fable 5
LMArena Text: writing and language15011405Claude Fable 5
LMArena Text: entertainment and media14821393Claude Fable 5
LMArena Vision: English13231267Claude Fable 5
LMArena Vision: Chinese13731326Claude Fable 5
LMArena Vision: reading text13371281Claude Fable 5
LMArena Vision: humor13411235Claude Fable 5
LMArena Vision: creative writing13381237Claude Fable 5
LMArena Document1496Not measuredOnly Claude Fable 5 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 Fable 5 and Qwen3.8 27B
BenchmarkClaude Fable 5Qwen3.8 27BLeader
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
Qwen3.8 27B$0.0007Fits in one request

Qwen3.8 27B costs less

Repository review

50K input + 3K output tokens

Claude Fable 5$0.650Fits in one request
Qwen3.8 27B$0.0060Fits in one request

Qwen3.8 27B costs less

Agent loop

200K input + 10K output tokens

Claude Fable 5$2.50Fits in one request
Qwen3.8 27B$0.021Fits in one request

Qwen3.8 27B 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 Qwen3.8 27B
SpecificationAnthropic Claude Fable 5Qwen3.8 27B
MakerAnthropicQwen
WeightsNo public weights listedOpen weights
ReleasedJune 9, 2026August 14, 2026
Context window1M1M
Input $/Mtok$10.00$0.04
Output $/Mtok$50.00$1.35
Coding index82.173.6
Independent measurements behind it22
Cost per solved task$2.16$1.26
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Fable 5 better than Qwen3.8 27B for coding?

Pick Claude Fable 5 for coding: it scores 82.1 on the Vibe Coding Index against 73.6 for Qwen3.8 27B. Pick Qwen3.8 27B when output tokens cost 37.0x less, or when a solved task costs $1.26 against $2.16.

Which is cheaper, Claude Fable 5 or Qwen3.8 27B?

Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Qwen3.8 27B costs $0.04 and $1.35. Per solved SWE-bench Verified task, Claude Fable 5 costs $2.16 and Qwen3.8 27B $1.26.

Which has the bigger context window, Claude Fable 5 or Qwen3.8 27B?

Neither: both take 1M tokens.

How many benchmarks is the Claude Fable 5 vs Qwen3.8 27B comparison based on?

82 benchmarks have a result for both models as of October 21, 2026. Claude Fable 5's coding index rests on 2 independent benchmarks and Qwen3.8 27B'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-fable-5-vs-qwen3-8-27b" title="Claude Fable 5 vs Qwen3.8 27B, 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-qwen3-8-27b">LLMBenchmark.ai, Claude Fable 5 vs Qwen3.8 27B</a></p>