Skip to main content
BenchLM
Advertise

Model comparison

GPT-5.3 Codex vs Step 3.7 Flash

Data verified

Head-to-head evidence from 17 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

66.69/100
Margin
15.9pts
← winning
50.76/100
2 category wins0 category wins

BenchAlign evidence: GPT-5.3 Codex supported; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. GPT-5.3 Codex and Step 3.7 Flash share 17 comparable benchmark results. 2 of 8 categories are comparable. 5 results are unique to GPT-5.3 Codex; 13 to Step 3.7 Flash.

Updated July 16, 2026
Shared results
17
GPT-5.3 Codex only
5
Step 3.7 Flash only
13
Comparable categories
2 / 8

Pick GPT-5.3 Codex if you want the stronger benchmark profile. Step 3.7 Flash only becomes the better choice if you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 17 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 is clearly ahead on the provisional aggregate, 82 to 57. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GPT-5.3 Codex's sharpest advantage is in coding, where it averages 67.2 against 56.3. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 77.3% to 59.5%.

GPT-5.3 Codex is also the more expensive model on tokens at $1.75 input / $14.00 output per 1M tokens, versus $0.20 input / $1.15 output per 1M tokens for Step 3.7 Flash. That is roughly 12.2x on output cost alone. GPT-5.3 Codex gives you the larger context window at 400K, compared with 256K for Step 3.7 Flash.

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 GPT-5.3 Codex and Step 3.7 Flash
CategoryGPT-5.3 CodexΔStep 3.7 Flash
CodingGPT-5.3 Codex67.2Margin← 10.9Step 3.7 Flash56.3
AgenticGPT-5.3 Codex71.4Margin← 5.0Step 3.7 Flash66.4

Decisive benchmark drivers

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

More
A · GPT-5.3 CodexB · Step 3.7 Flash

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 77.3%B 59.5%
    Winner: GPT-5.3 CodexΔ 17.8
    Terminal-Bench 2.0: GPT-5.3 Codex scored 77.3%; Step 3.7 Flash scored 59.5%. GPT-5.3 Codex wins this benchmark.

  2. SWE-bench Pro

    Coding
    Source ↗
    A 56.8%B 56.3%
    Winner: GPT-5.3 CodexΔ 0.5
    SWE-bench Pro: GPT-5.3 Codex scored 56.8%; Step 3.7 Flash scored 56.3%. GPT-5.3 Codex wins this benchmark.

Operational comparison

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

MetricGPT-5.3 CodexStep 3.7 FlashComparison
Input / output priceUSD per 1M tokensGPT-5.3 Codex$1.75 input / $14 outputStep 3.7 Flash$0.2 input / $1.15 outputStep 3.7 Flash has the lower combined listed price.
Generation speedtokens per secondGPT-5.3 Codex79 tok/sStep 3.7 FlashNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenGPT-5.3 Codex88.26 sStep 3.7 FlashNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensGPT-5.3 Codex400KStep 3.7 Flash256KGPT-5.3 Codex lists the larger context window.

Benchmark Deep Dive

AgenticGPT-5.3 Codex wins
BenchmarkGPT-5.3 CodexStep 3.7 FlashResult
Terminal-Bench 2.0Source 77.3%59.5%GPT-5.3 Codex leads
OSWorld-VerifiedSource 64.7%—Not comparable
τ²-bench resultsSource 86%98.5%Step 3.7 Flash leads
Gert LabsSource 57.47%51.57%GPT-5.3 Codex leads
JobBenchSource 33.7%—Not comparable
BrowseCompSource —75.8%Not comparable
DeepSearchQASource —92.8%Not comparable
GDPval-AASource —25.9%Not comparable
ToolathlonSource —49.5%Not comparable
Claw-EvalSource —67.1%Not comparable
HLE w/ toolsSource —47.2%Not comparable
AA Agentic IndexSource —21.5%Not comparable
GDPval-AASource —1017Not comparable
APEX-Agents-AASource —14.8%Not comparable
CodingGPT-5.3 Codex wins
BenchmarkGPT-5.3 CodexStep 3.7 FlashResult
SWE-bench VerifiedSource 85%—Not comparable
SWE-bench ProSource 56.8%56.3%GPT-5.3 Codex leads
SWE-RebenchSource 58.2%—Not comparable
Vibe Code BenchSource 61.77%—Not comparable
Terminal-Bench HardSource 53.0%35.6%GPT-5.3 Codex leads
AA-SciCodeSource 53.2%40.0%GPT-5.3 Codex leads
Terminal-Bench 2.0Source —59.5%Not comparable
AA Coding IndexSource —39.6%Not comparable
Reasoning
BenchmarkGPT-5.3 CodexStep 3.7 FlashResult
AA-LCRSource 74.0%63.7%GPT-5.3 Codex leads
CritPtSource 16.9%2.3%GPT-5.3 Codex leads
Knowledge
BenchmarkGPT-5.3 CodexStep 3.7 FlashResult
Artificial Analysis Intelligence IndexSource 44.3%30.3%GPT-5.3 Codex leads
AA-GPQA DiamondSource 91.5%80.9%GPT-5.3 Codex leads
AA-HLESource 39.9%19.9%GPT-5.3 Codex leads
AA-Omniscience IndexSource 9.9%-37.5%GPT-5.3 Codex leads
AA-Omniscience AccuracySource 51.8%25.4%GPT-5.3 Codex leads
AA-Omniscience Hallucination RateSource 86.9%84.4%Step 3.7 Flash leads
Multimodal
BenchmarkGPT-5.3 CodexStep 3.7 FlashResult
AA-MMMU-ProSource 78.5%75.3%GPT-5.3 Codex leads
Design Arena WebsiteSource 11971218Step 3.7 Flash leads
SimpleVQASource —79.2%Not comparable
V*Source —95.3%Not comparable
Inst. Following
BenchmarkGPT-5.3 CodexStep 3.7 FlashResult
AA-IFBenchSource 75.4%67.3%GPT-5.3 Codex leads
Frequently Asked Questions (3)

Which is better, GPT-5.3 Codex or Step 3.7 Flash?

GPT-5.3 Codex is ahead on BenchLM's provisional leaderboard, 82 to 57. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 77.3% and 59.5%.

Which is better for coding, GPT-5.3 Codex or Step 3.7 Flash?

GPT-5.3 Codex has the edge for coding in this comparison, averaging 67.2 versus 56.3. Inside this category, Terminal-Bench Hard is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, GPT-5.3 Codex or Step 3.7 Flash?

GPT-5.3 Codex has the edge for agentic tasks in this comparison, averaging 71.4 versus 66.4. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

OpenAI Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 16, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.