Skip to main content
BenchLM
Advertise

Model comparison

Claude Sonnet 4.6 vs GPT-5.3 Codex

Data verified

Head-to-head evidence from 20 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

65.07/100
Margin
1.6pts
winning →
66.69/100
1 category wins1 category wins

Public leaderboard positions: Claude Sonnet 4.6 #32 (Supported); GPT-5.3 Codex #26 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Sonnet 4.6 and GPT-5.3 Codex share 20 comparable benchmark results. 2 of 8 categories are comparable. 13 results are unique to Claude Sonnet 4.6; 1 to GPT-5.3 Codex.

Updated July 20, 2026
Shared results
20
Claude Sonnet 4.6 only
13
GPT-5.3 Codex only
1
Comparable categories
2 / 8

Pick GPT-5.3 Codex if you want the stronger benchmark profile. Claude Sonnet 4.6 only becomes the better choice if coding is the priority or you would rather avoid the extra latency and token burn of a reasoning model.

Confidence note. This is a partial-evidence comparison with 20 shared benchmark results across 6 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

GPT-5.3 Codex has the cleaner BenchAlign overall profile here, landing at 66.69 versus 65.07. It is a real lead, but still close enough that category-level strengths matter more than the headline number.

GPT-5.3 Codex's sharpest advantage is in agentic, where it averages 71.4 against 65.2. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 59.1% to 77.3%. Claude Sonnet 4.6 does hit back in coding, so the answer changes if that is the part of the workload you care about most.

Claude Sonnet 4.6 is also the more expensive model on tokens at $3.00 input / $15.00 output per 1M tokens, versus $1.75 input / $14.00 output per 1M tokens for GPT-5.3 Codex. GPT-5.3 Codex is the reasoning model in the pair, while Claude Sonnet 4.6 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. GPT-5.3 Codex gives you the larger context window at 400K, compared with 200K for Claude Sonnet 4.6.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Claude Sonnet 4.6 and GPT-5.3 Codex
CategoryClaude Sonnet 4.6ΔGPT-5.3 Codex
AgenticClaude Sonnet 4.665.2Margin→ 6.2GPT-5.3 Codex71.4
CodingClaude Sonnet 4.669.1Margin← 1.9GPT-5.3 Codex67.2
KnowledgeClaude Sonnet 4.666.0MarginNo overlapGPT-5.3 CodexNot measured
MathClaude Sonnet 4.626.4MarginNo overlapGPT-5.3 CodexNot measured
MultimodalClaude Sonnet 4.677.4MarginNo overlapGPT-5.3 CodexNot measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Claude Sonnet 4.6B · GPT-5.3 Codex

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 59.1%B 77.3%
    Winner: GPT-5.3 CodexΔ 18.2
    Terminal-Bench 2.0: Claude Sonnet 4.6 scored 59.1%; GPT-5.3 Codex scored 77.3%. GPT-5.3 Codex wins this benchmark.

  2. OSWorld-Verified

    Agentic
    Source ↗
    A 72.1%B 64.7%
    Winner: Claude Sonnet 4.6Δ 7.4
    OSWorld-Verified: Claude Sonnet 4.6 scored 72.1%; GPT-5.3 Codex scored 64.7%. Claude Sonnet 4.6 wins this benchmark.

  3. SWE-bench Verified

    Coding
    Source ↗
    A 79.6%B 85%
    Winner: GPT-5.3 CodexΔ 5.4
    SWE-bench Verified: Claude Sonnet 4.6 scored 79.6%; GPT-5.3 Codex scored 85%. GPT-5.3 Codex wins this benchmark.

  4. SWE-Rebench

    Coding
    Source ↗
    A 60.7%B 58.2%
    Winner: Claude Sonnet 4.6Δ 2.5
    SWE-Rebench: Claude Sonnet 4.6 scored 60.7%; GPT-5.3 Codex scored 58.2%. Claude Sonnet 4.6 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricClaude Sonnet 4.6GPT-5.3 CodexComparison
Input / output priceUSD per 1M tokensClaude Sonnet 4.6$3 input / $15 outputGPT-5.3 Codex$1.75 input / $14 outputGPT-5.3 Codex has the lower combined listed price.
Generation speedtokens per secondClaude Sonnet 4.644 tok/sGPT-5.3 Codex79 tok/sGPT-5.3 Codex has the higher measured throughput.
First-answer latencyseconds to first tokenClaude Sonnet 4.61.48 sGPT-5.3 Codex88.26 sClaude Sonnet 4.6 reaches the first token sooner.
Context windowmaximum listed tokensClaude Sonnet 4.6200KGPT-5.3 Codex400KGPT-5.3 Codex lists the larger context window.

Benchmark Deep Dive

AgenticGPT-5.3 Codex wins
BenchmarkClaude Sonnet 4.6GPT-5.3 CodexResult
Terminal-Bench 2.0Source 59.1%77.3%GPT-5.3 Codex leads
OSWorld-VerifiedSource 72.1%64.7%Claude Sonnet 4.6 leads
Claw-EvalSource 67.8%—Not comparable
CyberGymSource 65.2%—Not comparable
τ²-bench resultsSource 79.5%86%GPT-5.3 Codex leads
Gert LabsSource 62.92%57.47%Claude Sonnet 4.6 leads
OSWorld 2.0Source 8.3%—Not comparable
JobBenchSource 36.9%33.7%Claude Sonnet 4.6 leads
CodingClaude Sonnet 4.6 wins
BenchmarkClaude Sonnet 4.6GPT-5.3 CodexResult
SWE-bench VerifiedSource 79.6%85%GPT-5.3 Codex leads
SWE-RebenchSource 60.7%58.2%Claude Sonnet 4.6 leads
React Native EvalsSource 80.6%—Not comparable
Vibe Code BenchSource 51.48%61.77%GPT-5.3 Codex leads
cursorBench31Source 48.8%—Not comparable
AA-SciCodeSource 46.9%53.2%GPT-5.3 Codex leads
FrontierCode 1.1 MainSource 24.3%—Not comparable
SWE-bench ProSource —56.8%Not comparable
Reasoning
BenchmarkClaude Sonnet 4.6GPT-5.3 CodexResult
AA-LCRSource 57.7%74.0%GPT-5.3 Codex leads
CritPtSource 0.9%16.9%GPT-5.3 Codex leads
Knowledge
BenchmarkClaude Sonnet 4.6GPT-5.3 CodexResult
GPQASource 89.9%—Not comparable
SuperGPQASource 95%—Not comparable
MMLU-ProSource 79.2%—Not comparable
HLESource 49%—Not comparable
Artificial Analysis Intelligence IndexSource 35.9%44.3%GPT-5.3 Codex leads
AA-GPQA DiamondSource 79.9%91.5%GPT-5.3 Codex leads
AA-HLESource 13.2%39.9%GPT-5.3 Codex leads
AA-Omniscience IndexSource -2.9%9.9%GPT-5.3 Codex leads
AA-Omniscience AccuracySource 38.0%51.8%GPT-5.3 Codex leads
AA-Omniscience Hallucination RateSource 65.9%86.9%Claude Sonnet 4.6 leads
Math
BenchmarkClaude Sonnet 4.6GPT-5.3 CodexResult
FrontierMath v2 (Tiers 1-3)Source 32.400%—Not comparable
FrontierMath v2 (Tier 4)Source 8.300%—Not comparable
Multimodal
BenchmarkClaude Sonnet 4.6GPT-5.3 CodexResult
CharXivSource 77.4%—Not comparable
AA-MMMU-ProSource 70.6%78.5%GPT-5.3 Codex leads
Design Arena WebsiteSource 13171195Claude Sonnet 4.6 leads
Inst. Following
BenchmarkClaude Sonnet 4.6GPT-5.3 CodexResult
AA-IFBenchSource 41.2%75.4%GPT-5.3 Codex leads
Frequently Asked Questions (3)

Which is better, Claude Sonnet 4.6 or GPT-5.3 Codex?

GPT-5.3 Codex is ahead on BenchLM's BenchAlign leaderboard, 66.69 to 65.07. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 59.1% and 77.3%.

Which is better for coding, Claude Sonnet 4.6 or GPT-5.3 Codex?

Claude Sonnet 4.6 has the edge for coding in this comparison, averaging 69.1 versus 67.2. Inside this category, Vibe Code Bench is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Claude Sonnet 4.6 or GPT-5.3 Codex?

GPT-5.3 Codex has the edge for agentic tasks in this comparison, averaging 71.4 versus 65.2. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Anthropic Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 20, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.