Enter opens the highlighted result. Esc closes.

Gemini 3.7 Flash vs GPT-5.3-Codex

GPT-5.3-Codex leads Gemini 3.7 Flash on coding, 60.0 to 58.0 on the Vibe Coding Index, and Gemini 3.7 Flash costs 3.7x less per output token ($3.75 vs $14.00 per million). Updated October 22, 2026; 27 benchmarks have a result for both.

Replace Gemini 3.7 Flash Replace GPT-5.3-Codex
Gemini 3.7 FlashGoogle
Coding
58.0
Overall
65.7
Output $/M
$3.75
25
GPT-5.3-CodexOpenAI
Coding
60.0
Overall
No data
Output $/M
$14.00
2
vs boards won of 27 shared

Gemini 3.7 Flash or GPT-5.3-Codex: which to pick

Pick GPT-5.3-Codex for coding: it scores 60.0 on the Vibe Coding Index against 58.0 for Gemini 3.7 Flash, though on 3 benchmarks against 6.

Pick Gemini 3.7 Flash for output tokens 3.7x cheaper ($3.75 vs $14.00 per million) and a larger context window (1.05M against 400K tokens).

Model AGemini 3.7 FlashGoogle58.0/1006 independent benchmarks$1.78 per solved taskModel BGPT-5.3-CodexOpenAI60.0/1003 independent benchmarks$0.59 per solved task
Shared results
27
Gemini 3.7 Flash only
87
GPT-5.3-Codex only
7
Like-for-like categories
6 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Gemini 3.7 Flash 58.0 from 6 benchmarks, GPT-5.3-Codex 60.0 from 3 benchmarks.
Widest category
Agentic work, 22.3 points apart on average over 5 shared benchmarks.
Widest single gap
Design Arena: 3D: Gemini 3.7 Flash 1307 against GPT-5.3-Codex 1007, 37.5 points apart on a 0 to 100 scale.
Output price
Gemini 3.7 Flash $3.75 against GPT-5.3-Codex $14.00 per million output tokens, 3.7x apart.
Context window
Gemini 3.7 Flash takes 1.05M tokens, GPT-5.3-Codex 400K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGemini 3.7 FlashBarsGPT-5.3-CodexTies
Coding81
Intelligence20
Agentic work50
Reasoning20
Web design80
General01
All shared boards252
CodingLike-for-likeSix of these boards feed the Vibe Coding Index81
Coding benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
LiveBench coding80.5%Not measuredOnly Gemini 3.7 Flash measured
LMArena WebDev15911407Gemini 3.7 Flash
AA Coding Index71.5Not measuredOnly Gemini 3.7 Flash measured
LiveCodeBench (Vals AI)88.7%87.3%Gemini 3.7 Flash
SWE-bench Verified (Vals AI)80.8%78.0%Gemini 3.7 Flash
Terminal-Bench 4 (Vals AI)12.1%Not measuredOnly Gemini 3.7 Flash measured
Vibe Code Bench (Vals AI)70.4%61.8%Gemini 3.7 Flash
Code migration (Vals AI)34.8%Not measuredOnly Gemini 3.7 Flash measured
ProgramBench (Vals AI)0.0%Not measuredOnly Gemini 3.7 Flash measured
IOI (Vals AI)67.8%53.8%Gemini 3.7 Flash
Terminal-Bench 2.1 (Vals AI)77.5%Not measuredOnly Gemini 3.7 Flash measured
SkillsBench (Vals AI)65.9%Not measuredOnly Gemini 3.7 Flash measured
CyberBench (Vals AI)43.8%Not measuredOnly Gemini 3.7 Flash measured
SciCode59.8%Not measuredOnly Gemini 3.7 Flash measured
Terminal-Bench Hard (Artificial Analysis run)Not measured53.0%Only GPT-5.3-Codex measured
Terminal-BenchNot measured78.4%Only GPT-5.3-Codex measured
FrontierCode43.6%Not measuredOnly Gemini 3.7 Flash measured
FrontierSWE20.3%Not measuredOnly Gemini 3.7 Flash measured
WeirdMLNot measured79.3%Only GPT-5.3-Codex measured
ALE-Bench904.31655.2GPT-5.3-Codex
LMArena Text: coding1499Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: software and IT1498Not measuredOnly Gemini 3.7 Flash measured
LMArena WebDev: React15871395Gemini 3.7 Flash
LMArena WebDev: HTML15931422Gemini 3.7 Flash
LMArena WebDev: from an image15991443Gemini 3.7 Flash
SciCode13.8%Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 13 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
AA Intelligence Index39.632.5Gemini 3.7 Flash
Epoch Capabilities Index158.0157.1Gemini 3.7 Flash
Agentic workLike-for-likePublic boards, shown as published, kept out of the index50
Agentic work benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
Design Arena: agents building web apps12231032Gemini 3.7 Flash
Design Arena: agents building full-stack apps1208983Gemini 3.7 Flash
Design Arena: agents building mobile apps12291053Gemini 3.7 Flash
Design Arena: agents building Android apps12431084Gemini 3.7 Flash
Design Arena: agents building Godot games12641124Gemini 3.7 Flash
Design Arena: agents building games1235Not measuredOnly Gemini 3.7 Flash measured
Design Arena: spreadsheets1058Not measuredOnly Gemini 3.7 Flash measured
LiveBench research76.5%Not measuredOnly Gemini 3.7 Flash measured
LiveBench browser56.0%Not measuredOnly Gemini 3.7 Flash measured
LiveBench long agentic28.3%Not measuredOnly Gemini 3.7 Flash measured
SREBench (Vals AI)4.6%Not measuredOnly Gemini 3.7 Flash measured
Finance Agent (Vals AI)59.0%Not measuredOnly Gemini 3.7 Flash measured
Tax Agent Bench (Vals AI)22.8%Not measuredOnly Gemini 3.7 Flash measured
Legal Research Bench (Vals AI)34.6%Not measuredOnly Gemini 3.7 Flash measured
Excel Modeling Benchmark (Vals AI)71.3%Not measuredOnly Gemini 3.7 Flash measured
Terminal-Bench-Science (Vals AI)7.1%Not measuredOnly Gemini 3.7 Flash measured
τ²-Bench TelecomNot measured86.0%Only GPT-5.3-Codex measured
APEX-Agents67.8%Not measuredOnly Gemini 3.7 Flash measured
METR time horizonsNot measured74.5%Only GPT-5.3-Codex measured
Remote Labor Index5.0%Not measuredOnly Gemini 3.7 Flash measured
Vending-Bench 2Not measured$5,940Only GPT-5.3-Codex measured
Neither model has a result on the other 20 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index20
Reasoning benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
GPQA Diamond93.9%Not measuredOnly Gemini 3.7 Flash measured
LiveBench reasoning87.8%Not measuredOnly Gemini 3.7 Flash measured
MMMU Pro (Vals AI)89.0%Not measuredOnly Gemini 3.7 Flash measured
GPQA Diamond (Artificial Analysis run)94.5%91.5%Gemini 3.7 Flash
Humanity's Last Exam (Artificial Analysis run)47.9%42.5%Gemini 3.7 Flash
GPQA Diamond (Epoch AI run)94.8%Not measuredOnly Gemini 3.7 Flash measured
ARC-AGI95.5%Not measuredOnly Gemini 3.7 Flash measured
ARC-AGI-284.6%Not measuredOnly Gemini 3.7 Flash measured
Chess puzzles66.0%Not measuredOnly Gemini 3.7 Flash measured
Mystery game puzzles37.0%Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: hard prompts1495Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: expert1512Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: hard prompts in English1487Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: diagrams1329Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: homework1320Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 7 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Math benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
OTIS Mock AIME97.2%Not measuredOnly Gemini 3.7 Flash measured
FrontierMath, tiers 1 to 371.6%Not measuredOnly Gemini 3.7 Flash measured
FrontierMath, tier 436.6%Not measuredOnly Gemini 3.7 Flash measured
ProofBench58.0%Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: math1511Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: mathematical work1502Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 7 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index11 boards, none shared
Knowledge benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
MMLU Pro90.1%Not measuredOnly Gemini 3.7 Flash measured
LegalBench (Vals AI)87.3%Not measuredOnly Gemini 3.7 Flash measured
TaxEval (Vals AI)74.7%Not measuredOnly Gemini 3.7 Flash measured
MortgageTax (Vals AI)66.7%Not measuredOnly Gemini 3.7 Flash measured
MedCode (Vals AI)53.4%Not measuredOnly Gemini 3.7 Flash measured
SimpleQA Verified69.2%Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: science1511Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: medicine and healthcare1494Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: legal and government1496Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: business and finance1479Not measuredOnly Gemini 3.7 Flash measured
FACTS Benchmark Suite71.2%Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
Design Arena: websites13031162Gemini 3.7 Flash
Design Arena: UI components12961134Gemini 3.7 Flash
Design Arena: game dev13171161Gemini 3.7 Flash
Design Arena: data viz13151164Gemini 3.7 Flash
Design Arena: 3D13071007Gemini 3.7 Flash
Design Arena: all code categories13081144Gemini 3.7 Flash
Design Arena: SVG12831129Gemini 3.7 Flash
Design Arena: ASCII art12761152Gemini 3.7 Flash
LiveBench design73.2%Not measuredOnly Gemini 3.7 Flash measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index01
General benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
LiveBench data analysis68.0%Not measuredOnly Gemini 3.7 Flash measured
SAGE (Vals AI)49.2%Not measuredOnly Gemini 3.7 Flash measured
MedScribe (Vals AI)83.9%Not measuredOnly Gemini 3.7 Flash measured
IFBenchNot measured75.4%Only GPT-5.3-Codex measured
Long context reasoning (AA-LCR)83.0%83.3%GPT-5.3-Codex
DTBench96.8%Not measuredOnly Gemini 3.7 Flash measured
LMCA50.4%Not measuredOnly Gemini 3.7 Flash measured
Furniture assembly26.7%Not measuredOnly Gemini 3.7 Flash measured
LMArena Text1489Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: instruction following1484Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: creative writing1488Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision1314Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: English1483Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: non-English1487Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Chinese1552Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: German1506Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Japanese1511Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Korean1481Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: French1500Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Spanish1495Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Russian1515Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Polish1503Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: multi-turn1485Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: longer queries1492Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: writing and language1491Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: entertainment and media1477Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: English1309Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: Chinese1394Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: reading text1322Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: humor1299Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: creative writing1312Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 3.7 Flash and GPT-5.3-Codex
BenchmarkGemini 3.7 FlashGPT-5.3-CodexLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Gemini 3.7 Flash$0.0026Fits in one request
GPT-5.3-Codex$0.0088Fits in one request

Gemini 3.7 Flash costs less

Repository review

50K input + 3K output tokens

Gemini 3.7 Flash$0.049Fits in one request
GPT-5.3-Codex$0.130Fits in one request

Gemini 3.7 Flash costs less

Agent loop

200K input + 10K output tokens

Gemini 3.7 Flash$0.188Fits in one request
GPT-5.3-Codex$0.490Fits in one request

Gemini 3.7 Flash costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Gemini 3.7 Flash and GPT-5.3-Codex
SpecificationGoogle Gemini 3.7 FlashOpenAI GPT-5.3-Codex
MakerGoogleOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedAugust 13, 2026February 24, 2026
Context window1.05M400K
Input $/Mtok$0.75$1.75
Output $/Mtok$3.75$14.00
Coding index58.060.0
Independent measurements behind it63
Cost per solved task$1.78$0.59
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 3.7 Flash better than GPT-5.3-Codex for coding?

Pick GPT-5.3-Codex for coding: it scores 60.0 on the Vibe Coding Index against 58.0 for Gemini 3.7 Flash, though on 3 benchmarks against 6. Pick Gemini 3.7 Flash for output tokens 3.7x cheaper ($3.75 vs $14.00 per million) and a larger context window (1.05M against 400K tokens).

Which is cheaper, Gemini 3.7 Flash or GPT-5.3-Codex?

Gemini 3.7 Flash costs $0.75 per million input tokens and $3.75 per million output tokens; GPT-5.3-Codex costs $1.75 and $14.00. Per solved SWE-bench Verified task, Gemini 3.7 Flash costs $1.78 and GPT-5.3-Codex $0.59.

Which has the bigger context window, Gemini 3.7 Flash or GPT-5.3-Codex?

Gemini 3.7 Flash, with 1.05M tokens against 400K.

How many benchmarks is the Gemini 3.7 Flash vs GPT-5.3-Codex comparison based on?

27 benchmarks have a result for both models as of October 22, 2026. Gemini 3.7 Flash's coding index rests on 6 independent benchmarks and GPT-5.3-Codex's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/gemini-3-7-flash-vs-gpt-5-3-codex" title="Gemini 3.7 Flash vs GPT-5.3-Codex, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/gemini-3-7-flash-vs-gpt-5-3-codex">LLMBenchmark.ai, Gemini 3.7 Flash vs GPT-5.3-Codex</a></p>