Enter opens the highlighted result. Esc closes.

Claude Fable 5.2 vs Hy4 flash

Claude Fable 5.2 leads Hy4 flash on coding, 87.5 to 44.5 on the Vibe Coding Index, and Hy4 flash costs 284.7x less per output token ($0.1405 vs $40.00 per million). Updated November 22, 2026; 12 benchmarks have a result for both.

Claude Fable 5.2Anthropic
Coding
87.5
Overall
89.5
Output $/M
$40.00
11
Hy4 flashTencent
Coding
44.5
Overall
No data
Output $/M
$0.1405
1
vs boards won of 12 shared

Claude Fable 5.2 or Hy4 flash: which to pick

Pick Claude Fable 5.2 for coding: it scores 87.5 on the Vibe Coding Index against 44.5 for Hy4 flash.

Pick Hy4 flash for output tokens 284.7x cheaper ($0.1405 vs $40.00 per million) and a larger context window (1.05M against 1M tokens).

Shared results
12
Claude Fable 5.2 only
88
Hy4 flash only
0
Like-for-like categories
4 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 5.2 87.5 from 4 benchmarks, Hy4 flash 44.5 from 2 benchmarks.
Widest category
Coding, 44.8 points apart on average over 3 shared benchmarks.
Widest single gap
Terminal-Bench 4 (Vals AI): Claude Fable 5.2 65.2% against Hy4 flash 15.2%, 66.7 points apart on a 0 to 100 scale.
Output price
Claude Fable 5.2 $40.00 against Hy4 flash $0.1405 per million output tokens, 284.7x apart.
Context window
Claude Fable 5.2 takes 1M tokens, Hy4 flash 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 5.2BarsHy4 flashTies
Coding30
Agentic work21
Web design50
General10
All shared boards111
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
Coding benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
LiveBench coding79.2%Not measuredOnly Claude Fable 5.2 measured
LMArena WebDev1824Not measuredOnly Claude Fable 5.2 measured
Terminal-Bench 4 (Vals AI)65.2%15.2%Claude Fable 5.2
Vibe Code Bench (Vals AI)93.8%74.8%Claude Fable 5.2
ProgramBench (Vals AI)19.0%Not measuredOnly Claude Fable 5.2 measured
SWE-bench Multimodal (Vals AI)62.3%Not measuredOnly Claude Fable 5.2 measured
SciCode67.7%Not measuredOnly Claude Fable 5.2 measured
Terminal-Bench 4.0 (Artificial Analysis run)61.6%Not measuredOnly Claude Fable 5.2 measured
FrontierCode 2.049.1%Not measuredOnly Claude Fable 5.2 measured
FrontierSWE69.9%Not measuredOnly Claude Fable 5.2 measured
ALE-Bench2509.8Not measuredOnly Claude Fable 5.2 measured
LMArena Text: coding1487Not measuredOnly Claude Fable 5.2 measured
LMArena Text: software and IT1484Not measuredOnly Claude Fable 5.2 measured
LMArena WebDev: React1885Not measuredOnly Claude Fable 5.2 measured
LMArena WebDev: HTML1781Not measuredOnly Claude Fable 5.2 measured
GameBench 299.450.5Claude Fable 5.2
CursorBench 4.061.3%Not measuredOnly Claude Fable 5.2 measured
Neither model has a result on the other 35 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Intelligence benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
AA Intelligence Index64.6Not measuredOnly Claude Fable 5.2 measured
Epoch Capabilities Index173.4Not measuredOnly Claude Fable 5.2 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index21
Agentic work benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
Design Arena: agents building full-stack apps1411Not measuredOnly Claude Fable 5.2 measured
Design Arena: agents building mobile apps1320Not measuredOnly Claude Fable 5.2 measured
Design Arena: spreadsheets1421Not measuredOnly Claude Fable 5.2 measured
LiveBench research94.5%Not measuredOnly Claude Fable 5.2 measured
LiveBench browser86.9%Not measuredOnly Claude Fable 5.2 measured
LiveBench long agentic82.9%Not measuredOnly Claude Fable 5.2 measured
Tax Agent Bench (Vals AI)32.1%23.3%Claude Fable 5.2
Legal Research Bench (Vals AI)46.1%38.5%Claude Fable 5.2
Harvey's Legal Agent Benchmark (Vals AI)5.0%5.4%Hy4 flash
BioMysteryBench (Vals AI)83.7%Not measuredOnly Claude Fable 5.2 measured
Terminal-Bench-Science (Vals AI)55.7%Not measuredOnly Claude Fable 5.2 measured
Vals Science Index63.7%Not measuredOnly Claude Fable 5.2 measured
APEX-Agents80.1%Not measuredOnly Claude Fable 5.2 measured
Remote Labor Index41.3%Not measuredOnly Claude Fable 5.2 measured
PostTrainBench44.6%Not measuredOnly Claude Fable 5.2 measured
SpireClimb15.3%Not measuredOnly Claude Fable 5.2 measured
Vending-Bench 2$9,628Not measuredOnly Claude Fable 5.2 measured
Neither model has a result on the other 27 boards here
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index19 boards, none shared
Reasoning benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
LiveBench reasoning88.8%Not measuredOnly Claude Fable 5.2 measured
MysteryMechanism (Vals AI)51.8%Not measuredOnly Claude Fable 5.2 measured
Humanity's Last Exam (Artificial Analysis run)74.3%Not measuredOnly Claude Fable 5.2 measured
HLE-Diamond (Artificial Analysis run)80.2%Not measuredOnly Claude Fable 5.2 measured
CritPt v2, hard subset (Artificial Analysis run)76.5%Not measuredOnly Claude Fable 5.2 measured
GPQA Diamond (Epoch AI run)93.9%Not measuredOnly Claude Fable 5.2 measured
ARC-AGI98.5%Not measuredOnly Claude Fable 5.2 measured
ARC-AGI-293.8%Not measuredOnly Claude Fable 5.2 measured
SimpleBench94.9%Not measuredOnly Claude Fable 5.2 measured
Chess puzzles79.0%Not measuredOnly Claude Fable 5.2 measured
Mystery game puzzles90.0%Not measuredOnly Claude Fable 5.2 measured
EBR-bench90.5%Not measuredOnly Claude Fable 5.2 measured
LMArena Text: hard prompts1484Not measuredOnly Claude Fable 5.2 measured
LMArena Text: expert1510Not measuredOnly Claude Fable 5.2 measured
LMArena Text: hard prompts in English1483Not measuredOnly Claude Fable 5.2 measured
LMArena Vision: diagrams1322Not measuredOnly Claude Fable 5.2 measured
LMArena Vision: homework1354Not measuredOnly Claude Fable 5.2 measured
Humanity's Last Exam (Scale run)72.4%Not measuredOnly Claude Fable 5.2 measured
GoBench2564Not measuredOnly Claude Fable 5.2 measured
Neither model has a result on the other 7 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Math benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
OTIS Mock AIME99.7%Not measuredOnly Claude Fable 5.2 measured
FrontierMath, tiers 1 to 395.1%Not measuredOnly Claude Fable 5.2 measured
FrontierMath, tier 4100.0%Not measuredOnly Claude Fable 5.2 measured
ProofBench100.0%Not measuredOnly Claude Fable 5.2 measured
LMArena Text: math1491Not measuredOnly Claude Fable 5.2 measured
LMArena Text: mathematical work1504Not measuredOnly Claude Fable 5.2 measured
MathArena82.1%Not measuredOnly Claude Fable 5.2 measured
Neither model has a result on the other 7 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Knowledge benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
AA-Omniscience66.8Not measuredOnly Claude Fable 5.2 measured
SimpleQA Verified94.0%Not measuredOnly Claude Fable 5.2 measured
LMArena Text: science1494Not measuredOnly Claude Fable 5.2 measured
LMArena Text: medicine and healthcare1464Not measuredOnly Claude Fable 5.2 measured
LMArena Text: legal and government1472Not measuredOnly Claude Fable 5.2 measured
LMArena Text: business and finance1443Not measuredOnly Claude Fable 5.2 measured
PRBench Finance47.1%Not measuredOnly Claude Fable 5.2 measured
PRBench Legal47.1%Not measuredOnly Claude Fable 5.2 measured
Neither model has a result on the other 13 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index50
Web design benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
Design Arena: websites13481227Claude Fable 5.2
Design Arena: UI components13771262Claude Fable 5.2
Design Arena: game dev15361202Claude Fable 5.2
Design Arena: 3D14921209Claude Fable 5.2
Design Arena: all code categories13921225Claude Fable 5.2
LiveBench design85.1%Not measuredOnly Claude Fable 5.2 measured
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index10
General benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
Vals Index71.3%50.8%Claude Fable 5.2
LiveBench data analysis81.5%Not measuredOnly Claude Fable 5.2 measured
Long context reasoning (AA-LCR)83.3%Not measuredOnly Claude Fable 5.2 measured
GDPval-AA1831Not measuredOnly Claude Fable 5.2 measured
DTBench99.2%Not measuredOnly Claude Fable 5.2 measured
LMCA70.6%Not measuredOnly Claude Fable 5.2 measured
Furniture assembly90.0%Not measuredOnly Claude Fable 5.2 measured
LMArena Text1478Not measuredOnly Claude Fable 5.2 measured
LMArena Text: instruction following1490Not measuredOnly Claude Fable 5.2 measured
LMArena Text: creative writing1516Not measuredOnly Claude Fable 5.2 measured
LMArena Vision1313Not measuredOnly Claude Fable 5.2 measured
LMArena Text: English1474Not measuredOnly Claude Fable 5.2 measured
LMArena Text: non-English1474Not measuredOnly Claude Fable 5.2 measured
LMArena Text: Chinese1528Not measuredOnly Claude Fable 5.2 measured
LMArena Text: French1524Not measuredOnly Claude Fable 5.2 measured
LMArena Text: Spanish1446Not measuredOnly Claude Fable 5.2 measured
LMArena Text: Russian1501Not measuredOnly Claude Fable 5.2 measured
LMArena Text: multi-turn1458Not measuredOnly Claude Fable 5.2 measured
LMArena Text: longer queries1499Not measuredOnly Claude Fable 5.2 measured
LMArena Text: writing and language1498Not measuredOnly Claude Fable 5.2 measured
LMArena Text: entertainment and media1486Not measuredOnly Claude Fable 5.2 measured
LMArena Vision: English1315Not measuredOnly Claude Fable 5.2 measured
LMArena Vision: reading text1327Not measuredOnly Claude Fable 5.2 measured
MultiChallenge80.4%Not measuredOnly Claude Fable 5.2 measured
Neither model has a result on the other 25 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Fable 5.2 and Hy4 flash
BenchmarkClaude Fable 5.2Hy4 flashLeader
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 5.2$0.028Fits in one request
Hy4 flash$0.0001Fits in one request

Hy4 flash costs less

Repository review

50K input + 3K output tokens

Claude Fable 5.2$0.520Fits in one request
Hy4 flash$0.0022Fits in one request

Hy4 flash costs less

Agent loop

200K input + 10K output tokens

Claude Fable 5.2$2.00Fits in one request
Hy4 flash$0.0084Fits in one request

Hy4 flash 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 5.2 and Hy4 flash
SpecificationAnthropic Claude Fable 5.2Tencent Hy4 flash
MakerAnthropicTencent
WeightsNo public weights listedOpen weights
ReleasedNovember 9, 2026November 16, 2026
Context window1M1.05M
Input $/Mtok$8.00$0.035
Output $/Mtok$40.00$0.1405
Coding index87.544.5
Independent measurements behind it42
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Fable 5.2 better than Hy4 flash for coding?

Pick Claude Fable 5.2 for coding: it scores 87.5 on the Vibe Coding Index against 44.5 for Hy4 flash. Pick Hy4 flash for output tokens 284.7x cheaper ($0.1405 vs $40.00 per million) and a larger context window (1.05M against 1M tokens).

Which is cheaper, Claude Fable 5.2 or Hy4 flash?

Claude Fable 5.2 costs $8.00 per million input tokens and $40.00 per million output tokens; Hy4 flash costs $0.035 and $0.1405. Cost per solved task is known for neither of them.

Which has the bigger context window, Claude Fable 5.2 or Hy4 flash?

Hy4 flash, with 1.05M tokens against 1M.

How many benchmarks is the Claude Fable 5.2 vs Hy4 flash comparison based on?

12 benchmarks have a result for both models as of November 22, 2026. Claude Fable 5.2's coding index rests on 4 independent benchmarks and Hy4 flash'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-5-2-vs-tencent-hy4-flash" title="Claude Fable 5.2 vs Hy4 flash, 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-5-2-vs-tencent-hy4-flash">LLMBenchmark.ai, Claude Fable 5.2 vs Hy4 flash</a></p>