Noometry

Model comparison

Claude Fable 5 vs Qwen2.5 72B Instruct

Claude Fable 5 is the stronger model overall, scoring 66.3 to 31.5 on the Noometry Index. Qwen2.5 72B Instruct costs 8.2× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

Last verified . 23 shared benchmarks.

Claude Fable 5 Anthropic

66.3

Rank #5 Confirmed

Qwen2.5 72B Instruct Alibaba (Qwen)

31.5

Rank #259 Confirmed

Summary

  • They share 23 benchmarks with published results for both. Claude Fable 5 scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Fable 5 leads 88.4 to 18.1.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for Claude Fable 5 and 8.1% for Qwen2.5 72B Instruct.
  • Qwen2.5 72B Instruct is cheaper at $1.40 / $5.60 per million input/output tokens, against $10 / $50 for Claude Fable 5.
  • Claude Fable 5 accepts more context: 1M tokens versus 131K.
  • Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

Side by side

Claude Fable 5Qwen2.5 72B Instruct
ProviderAnthropicAlibaba (Qwen)
Noometry Index66.331.5
Released2026-06-072024-09
WeightsProprietaryOpen
Context window1M131K
Max output128K8K
Input $ / M tokens$10$1.40
Output $ / M tokens$50$5.60
Results tracked6143

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Claude Fable 5 leads

Claude Fable 5: 67.0 (#4), Qwen2.5 72B Instruct: 32.8 (#256)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
WeirdML91.9%16%
LMArena Coding15191293
DeepSWE69.9%—
FrontierCode53.5%—
LMArena WebDev1626—
FrontierSWE47%—
SciCode61%—
BigCodeBench Instruct—45.8%
MirrorCode63.9%—
BigCodeBench Complete—55.9%
ALE-Bench2,041—

Agentic & Tool Use Claude Fable 5 leads

Claude Fable 5: 54.1 (#2), Qwen2.5 72B Instruct: 22.0 (#124)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
APEX-Agents63.6%—
Remote Labor Index16.1%—
TheAgentCompany—5.7%
τ²-bench Banking39.7%—
PostTrainBench41.8%—
BALROG—16.2%
GBAEval74.5%—
GDP.pdf30%—
LMArena Search1230—
METR Time Horizons—35.8%
Vending-Bench 25,680—

Reasoning Claude Fable 5 leads

Claude Fable 5: 76.8 (#4), Qwen2.5 72B Instruct: 21.4 (#197)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
LMArena Hard Prompts15081271
DTBench98.4%62.9%
LMCA61.1%13.4%
Epoch Capabilities Index163.6129.01
ARC-AGI-289.2%—
SimpleBench81.9%—
Kagi LLM Benchmark91.4%—
NYT Connections (extended)92.7%—
ARC-AGI-198.5%—
CritPt28.6%—
Chess Puzzles41%—
EnigmaEval39.3%—
EBR-Bench39.5%—
Mystery Game Puzzles52%—
Surface Evolver Bench95%—
Bench to the Future 30.13—
BIG-Bench Hard—79.8%
ForecastBench—57.5
HellaSwag—84.8%
PIQA—82.6%
WinoGrande—82.3%

Math Claude Fable 5 leads

Claude Fable 5: 88.4 (#5), Qwen2.5 72B Instruct: 18.1 (#281)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
OTIS Mock AIME 2024-2025100%8.1%
LMArena Math15191283
FrontierMath (Tiers 1-3)87%—
FrontierMath Tier 490.2%—
ProofBench95%—
Omni-MATH—33%
MATH Level 5—63.2%
FrontierMath Erdős0%—

Knowledge Claude Fable 5 leads

Claude Fable 5: 62.3 (#22), Qwen2.5 72B Instruct: 26.9 (#247)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
GPQA Diamond85.9%49.1%
LMArena Expert15341245
SimpleQA Verified70.7%—
MMLU-Pro—63.1%
Confabulations—19.1%
GPQA (HELM)—42.6%
ARC (AI2) Challenge—94.5%
MMLU—85.3%
TriviaQA—71.9%

Multimodal Not comparable

Claude Fable 5: 45.6 (#14), Qwen2.5 72B Instruct: —

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
LMArena Vision1326—
Blueprint-Bench 238.6%—
Furniture Assembly35.8%—
LMArena Document1496—

Multilingual Claude Fable 5 leads

Claude Fable 5: 57.6 (#8), Qwen2.5 72B Instruct: 41.2 (#208)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
LMArena Non-English14831252
LMArena Chinese15461272
LMArena French15091279
LMArena German14861234
LMArena Japanese15071180
LMArena Korean14931188
LMArena Russian15071264
LMArena Spanish15031255

Instruction Following Claude Fable 5 leads

Claude Fable 5: 78.8 (#7), Qwen2.5 72B Instruct: 65.7 (#216)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
LMArena Instruction Following15031254
IFEval—80.6%

Long Context Claude Fable 5 leads

Claude Fable 5: 46.3 (#22), Qwen2.5 72B Instruct: 38.9 (#183)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
LMArena Longer Query15101282

Writing & Preference Claude Fable 5 leads

Claude Fable 5: 76.1 (#5), Qwen2.5 72B Instruct: 46.9 (#210)

BenchmarkClaude Fable 5Qwen2.5 72B Instruct
LMArena Text14911269
LMArena Creative Writing14941221
LMArena Multi-Turn15051272
EQ-Bench Creative Writing1943—
WildBench—80.2%
EQ-Bench 41340—

Frequently asked questions

Is Claude Fable 5 better than Qwen2.5 72B Instruct?

Claude Fable 5 is the stronger model overall, scoring 66.3 to 31.5 on the Noometry Index. Qwen2.5 72B Instruct costs 8.2× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

Which is cheaper, Claude Fable 5 or Qwen2.5 72B Instruct?

Qwen2.5 72B Instruct is cheaper. It lists at $1.40 per million input tokens and $5.60 per million output tokens; Claude Fable 5 lists at $10 and $50.

Is Claude Fable 5 or Qwen2.5 72B Instruct better for coding?

Claude Fable 5 scores higher on coding benchmarks: 67.0 versus 32.8 in the Noometry coding category.

Which has the bigger context window?

Claude Fable 5 does, with 1M tokens against 131K.

How many benchmarks do Claude Fable 5 and Qwen2.5 72B Instruct share?

23 benchmarks have published results for both models. Claude Fable 5 has 61 scored results on Noometry and Qwen2.5 72B Instruct has 43.

Related comparisons

Go deeper