Noometry

Model comparison

Claude Opus 4.5 vs Phi-4

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 31.2 on the Noometry Index. Phi-4 costs 114× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

Last verified . 24 shared benchmarks.

Claude Opus 4.5 Anthropic

50.5

Rank #47 Confirmed

Phi-4 Microsoft

31.2

Rank #279 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Claude Opus 4.5 scores higher in 9 categories and Phi-4 in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Claude Opus 4.5 leads 68.1 to 40.5.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.1% for Claude Opus 4.5 and 13.8% for Phi-4.
  • Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
  • Claude Opus 4.5 accepts more context: 200K tokens versus 128K.
  • Phi-4 has downloadable open weights; the other is API-only.

Side by side

Claude Opus 4.5Phi-4
ProviderAnthropicMicrosoft
Noometry Index50.531.2
Released2025-11-012024-12-11
WeightsProprietaryOpen
Context window200K128K
Max output64K4K
Input $ / M tokens$5$0.07
Output $ / M tokens$25$0.14
Results tracked6937

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

Category by category

Coding Claude Opus 4.5 leads

Claude Opus 4.5: 54.8 (#27), Phi-4: 34.4 (#239)

BenchmarkClaude Opus 4.5Phi-4
LMArena Coding15041231
SWE-bench Verified76.7%—
SWE-bench Verified (bash only)76.8%—
LMArena WebDev1494—
SWE-bench Multilingual70.7%—
GSO26.5%—
WeirdML63.7%—
BigCodeBench Instruct—45.5%
LiveBench Coding—30.7%
BigCodeBench Complete—55.4%
ALE-Bench1,025—
AlgoTune1.77—

Agentic & Tool Use Claude Opus 4.5 leads

Claude Opus 4.5: 47.3 (#12), Phi-4: 22.8 (#128)

BenchmarkClaude Opus 4.5Phi-4
Berkeley Function Calling Leaderboard77.5%28.8%
BALROG43.5%11.6%
Terminal-Bench63.1%—
GDPval45.5%—
Remote Labor Index3.8%—
τ²-bench Airline84%—
τ²-bench Banking24.7%—
τ²-bench Retail79.6%—
τ²-bench Telecom92.3%—
Cybench82%—
DeepResearch Bench54.8%—
OSWorld66.3%—
LMArena Search1180—
METR Time Horizons75%—
Vending-Bench 24,967—

Reasoning Claude Opus 4.5 leads

Claude Opus 4.5: 42.6 (#51), Phi-4: 17.7 (#291)

BenchmarkClaude Opus 4.5Phi-4
Chess Puzzles12%1%
LMArena Hard Prompts14761220
Epoch Capabilities Index150.09130.42
ARC-AGI-237.6%—
SimpleBench62%—
Kagi LLM Benchmark80.2%—
NYT Connections (extended)52.5%—
ARC-AGI-180%—
EnigmaEval11.9%—
EBR-Bench14.3%—
LiveBench Reasoning—47.8%
Mystery Game Puzzles22%—
DTBench89.9%—
LiveBench Data Analysis—45.2%
LMCA44.5%—
ForecastBench60.7—
LiveBench—41.6%

Math Claude Opus 4.5 leads

Claude Opus 4.5: 38.6 (#132), Phi-4: 20.8 (#285)

BenchmarkClaude Opus 4.5Phi-4
OTIS Mock AIME 2024-202586.1%13.8%
LMArena Math14631246
FrontierMath (Tiers 1-3)34.4%—
FrontierMath Tier 44.9%—
ProofBench36%—
LiveBench Math—42%
MATH Level 5—64.9%
FrontierMath (Feb 2025 set)20.7%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Claude Opus 4.5 leads

Claude Opus 4.5: 56.5 (#44), Phi-4: 32.6 (#209)

BenchmarkClaude Opus 4.5Phi-4
GPQA Diamond86%56.1%
Vectara Hallucination Rate10.9%3.7%
LMArena Expert14871203
Humanity's Last Exam25.2%—
SimpleQA Verified45.7%—
Confabulations—29.4%
MMLU—84.8%

Multimodal Not comparable

Claude Opus 4.5: 31.4 (#107), Phi-4: —

BenchmarkClaude Opus 4.5Phi-4
GeoBench75%—
VPCT40%—
Furniture Assembly28.3%—
LMArena Document1462—

Multilingual Claude Opus 4.5 leads

Claude Opus 4.5: 54.3 (#47), Phi-4: 37.2 (#237)

BenchmarkClaude Opus 4.5Phi-4
LMArena Non-English14381197
LMArena Chinese14701212
LMArena French14711224
LMArena German14491222
LMArena Japanese14161158
LMArena Korean14241151
LMArena Russian14471209
LMArena Spanish14581234

Instruction Following Claude Opus 4.5 leads

Claude Opus 4.5: 77.5 (#19), Phi-4: 60.4 (#251)

BenchmarkClaude Opus 4.5Phi-4
LMArena Instruction Following14781201
LiveBench Instruction Following—58.4%

Long Context Claude Opus 4.5 leads

Claude Opus 4.5: 46.5 (#22), Phi-4: 36.9 (#226)

BenchmarkClaude Opus 4.5Phi-4
LMArena Longer Query14801217
CL-bench21.1%—

Writing & Preference Claude Opus 4.5 leads

Claude Opus 4.5: 68.1 (#28), Phi-4: 40.5 (#244)

BenchmarkClaude Opus 4.5Phi-4
LMArena Text14511217
LMArena Creative Writing14451182
LMArena Multi-Turn14661206
Short-Story Creative Writing—62.6%
EQ-Bench Creative Writing1687—
LiveBench Language—25.6%

Frequently asked questions

Is Claude Opus 4.5 better than Phi-4?

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 31.2 on the Noometry Index. Phi-4 costs 114× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 4.5 or Phi-4?

Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

Is Claude Opus 4.5 or Phi-4 better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 34.4 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.5 does, with 200K tokens against 128K.

How many benchmarks do Claude Opus 4.5 and Phi-4 share?

24 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and Phi-4 has 37.

Related comparisons

Go deeper