Noometry

Model comparison

Claude Opus 5 vs Qwen3 235B-A22B

Claude Opus 5 is the stronger model overall, scoring 67.7 to 43.2 on the Noometry Index. Qwen3 235B-A22B costs 8.2× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

Last verified . 33 shared benchmarks.

Claude Opus 5 Anthropic

67.7

Rank #4 Confirmed

Qwen3 235B-A22B Alibaba (Qwen)

43.2

Rank #83 Confirmed

Summary

  • They share 33 benchmarks with published results for both. Claude Opus 5 scores higher in 9 categories and Qwen3 235B-A22B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Claude Opus 5 leads 77.1 to 15.8.
  • The biggest single-benchmark swing is ARC-AGI-2: 90.4% for Claude Opus 5 and 1.3% for Qwen3 235B-A22B.
  • Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $5 / $25 for Claude Opus 5.
  • Claude Opus 5 accepts more context: 1M tokens versus 131K.
  • Qwen3 235B-A22B has downloadable open weights; the other is API-only.

Side by side

Claude Opus 5Qwen3 235B-A22B
ProviderAnthropicAlibaba (Qwen)
Noometry Index67.743.2
Released2026-07-242025-04
WeightsProprietaryOpen
Context window1M131K
Max output128K16K
Input $ / M tokens$5$0.70
Output $ / M tokens$25$2.80
Results tracked5749

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

Category by category

Coding Claude Opus 5 leads

Claude Opus 5: 67.0 (#5), Qwen3 235B-A22B: 43.8 (#72)

BenchmarkClaude Opus 5Qwen3 235B-A22B
SciCode56.4%42.4%
WeirdML91.8%41%
LMArena Coding15331445
DeepSWE73.6%—
FrontierCode53.4%—
Aider Polyglot—59.6%
CursorBench46.6%—
LMArena WebDev1694—
FrontierSWE52%—
ALE-Bench2,165—

Agentic & Tool Use Claude Opus 5 leads

Claude Opus 5: 55.5 (#1), Qwen3 235B-A22B: 33.7 (#48)

BenchmarkClaude Opus 5Qwen3 235B-A22B
Vending-Bench 211,182-11.34
APEX-Agents65.8%—
Berkeley Function Calling Leaderboard—52.1%
OSWorld 2.031.4%—
τ²-bench Banking48.7%—
PostTrainBench35%—
BALROG63.4%—
GBAEval79.6%—
GDP.pdf24%—

Reasoning Claude Opus 5 leads

Claude Opus 5: 77.1 (#3), Qwen3 235B-A22B: 15.8 (#302)

BenchmarkClaude Opus 5Qwen3 235B-A22B
ARC-AGI-290.4%1.3%
SimpleBench80.6%31%
ARC-AGI-197.5%11%
CritPt29.1%0%
Chess Puzzles42%12%
LMArena Hard Prompts15251433
Mystery Game Puzzles59%9%
DTBench97.9%80.3%
LMCA64.5%29.3%
Epoch Capabilities Index162.67143.87
Kagi LLM Benchmark—69.4%
NYT Connections (extended)94.3%—
EBR-Bench45.7%—
Bench to the Future 30.12—
ForecastBench—59.7

Math Claude Opus 5 leads

Claude Opus 5: 86.1 (#6), Qwen3 235B-A22B: 48.5 (#51)

BenchmarkClaude Opus 5Qwen3 235B-A22B
OTIS Mock AIME 2024-202598.9%86.7%
LMArena Math15331431
FrontierMath (Tiers 1-3)85.6%—
FrontierMath Tier 473.2%—
ProofBench99%—
Omni-MATH—71.8%
MATH Level 5—68.9%
FrontierMath (Feb 2025 set)—8.5%
FrontierMath Tier 4 (v1)—0%

Knowledge Claude Opus 5 leads

Claude Opus 5: 66.8 (#8), Qwen3 235B-A22B: 49.5 (#67)

BenchmarkClaude Opus 5Qwen3 235B-A22B
GPQA Diamond93.9%80.1%
SimpleQA Verified59.9%40.4%
LMArena Expert15581460
MMLU-Pro—84.4%
Confabulations—15.6%
Vectara Hallucination Rate—9.3%
GPQA (HELM)—72.7%

Multimodal Not comparable

Claude Opus 5: 51.1 (#6), Qwen3 235B-A22B: —

BenchmarkClaude Opus 5Qwen3 235B-A22B
LMArena Vision1321—
Blueprint-Bench 230.4%—
Furniture Assembly60.8%—
LMArena Document1516—

Multilingual Claude Opus 5 leads

Claude Opus 5: 58.8 (#3), Qwen3 235B-A22B: 52.5 (#84)

BenchmarkClaude Opus 5Qwen3 235B-A22B
LMArena Non-English15001410
LMArena Chinese15751481
LMArena French15221445
LMArena German15221432
LMArena Japanese15101400
LMArena Korean15261391
LMArena Russian15071411
LMArena Spanish15191429

Instruction Following Claude Opus 5 leads

Claude Opus 5: 79.3 (#5), Qwen3 235B-A22B: 72.8 (#132)

BenchmarkClaude Opus 5Qwen3 235B-A22B
LMArena Instruction Following15161408
IFEval—83.5%

Long Context Too close to call

Claude Opus 5: 46.5 (#20), Qwen3 235B-A22B: 46.1 (#25)

BenchmarkClaude Opus 5Qwen3 235B-A22B
LMArena Longer Query15161426
Fiction.LiveBench—75%

Writing & Preference Claude Opus 5 leads

Claude Opus 5: 79.3 (#2), Qwen3 235B-A22B: 59.8 (#103)

BenchmarkClaude Opus 5Qwen3 235B-A22B
LMArena Text15061419
LMArena Creative Writing14911384
EQ-Bench Creative Writing21331366
LMArena Multi-Turn14961432
Short-Story Creative Writing—83%
WildBench—86.6%
EQ-Bench 41385—

Frequently asked questions

Is Claude Opus 5 better than Qwen3 235B-A22B?

Claude Opus 5 is the stronger model overall, scoring 67.7 to 43.2 on the Noometry Index. Qwen3 235B-A22B costs 8.2× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 5 or Qwen3 235B-A22B?

Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Claude Opus 5 lists at $5 and $25.

Is Claude Opus 5 or Qwen3 235B-A22B better for coding?

Claude Opus 5 scores higher on coding benchmarks: 67.0 versus 43.8 in the Noometry coding category.

Which has the bigger context window?

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

How many benchmarks do Claude Opus 5 and Qwen3 235B-A22B share?

33 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Qwen3 235B-A22B has 49.

Related comparisons

Go deeper