GPT-5.3-Codex-Spark Benchmark Scores & Performance

Benchmark analysis of GPT-5.3-Codex-Spark by OpenAI across 32 sourced tests on BenchLM.

According to BenchLM.ai, GPT-5.3-Codex-Spark ranks #7 out of 123 models with an overall score of 87/100. This places it in the upper tier of AI models, with competitive scores across most benchmark categories.

GPT-5.3-Codex-Spark is a proprietary model with a 256K token context window. It uses explicit chain-of-thought reasoning, which typically improves performance on math and complex reasoning tasks at the cost of higher latency and token usage.

GPT-5.3-Codex-Spark sits inside the GPT-5.3 Codex family alongside GPT-5.3 Codex.

Its strongest category is Coding (#5), while its weakest is Multimodal & Grounded (#18). This performance profile makes it particularly well-suited for software development and code generation tasks.

Creator

OpenAI

Source Type

Proprietary

Reasoning

Reasoning

Context Window

256K

Overall Score

87#7 of 123

Arena Elo

1398

Family & Lineage

Family

GPT-5.3 Codex

Spark

Canonical Entry

GPT-5.3 Codex

Sibling Models

Knowledge Benchmarks

MMLU
97
GPQA
95
SuperGPQA
93
OpenBookQA
91
MMLU-Pro
88
HLE
42
FrontierScience
88

Coding Benchmarks

HumanEval
91
SWE-bench Verified
80
LiveCodeBench
80
SWE-bench Pro
85

Mathematics Benchmarks

AIME 2023
98
AIME 2024
98
AIME 2025
97
HMMT Feb 2023
94
HMMT Feb 2024
96
HMMT Feb 2025
95
BRUMO 2025
95
MATH-500
98

Reasoning Benchmarks

SimpleQA
94
MuSR
92
BBH
97
LongBench v2
91
MRCRv2
92

Agentic Benchmarks

Terminal-Bench 2.0
90
BrowseComp
82
OSWorld-Verified
83

Multimodal & Grounded Benchmarks

MMMU-Pro
86
OfficeQA Pro
91

Instruction Following Benchmarks

IFEval
92

Multilingual Benchmarks

MGSM
94
MMLU-ProX
89

Frequently Asked Questions

How does GPT-5.3-Codex-Spark perform overall in AI benchmarks?

GPT-5.3-Codex-Spark ranks #7 out of 123 models with an overall score of 87. It is created by OpenAI and features a 256K context window.

Is GPT-5.3-Codex-Spark good for knowledge and understanding?

GPT-5.3-Codex-Spark ranks #11 out of 123 models in knowledge and understanding benchmarks with an average score of 78.3. There are stronger options in this category.

Is GPT-5.3-Codex-Spark good for coding and programming?

GPT-5.3-Codex-Spark ranks #5 out of 123 models in coding and programming benchmarks with an average score of 82.3. It is among the top performers in this category.

Is GPT-5.3-Codex-Spark good for mathematics?

GPT-5.3-Codex-Spark ranks #11 out of 123 models in mathematics benchmarks with an average score of 96.7. There are stronger options in this category.

Is GPT-5.3-Codex-Spark good for reasoning and logic?

GPT-5.3-Codex-Spark ranks #10 out of 123 models in reasoning and logic benchmarks with an average score of 92.7. It is among the top performers in this category.

Is GPT-5.3-Codex-Spark good for agentic tool use and computer tasks?

GPT-5.3-Codex-Spark ranks #7 out of 123 models in agentic tool use and computer tasks benchmarks with an average score of 85.6. It is among the top performers in this category.

Is GPT-5.3-Codex-Spark good for multimodal and grounded tasks?

GPT-5.3-Codex-Spark ranks #18 out of 123 models in multimodal and grounded tasks benchmarks with an average score of 88.3. There are stronger options in this category.

Is GPT-5.3-Codex-Spark good for instruction following?

GPT-5.3-Codex-Spark ranks #14 out of 123 models in instruction following benchmarks with an average score of 92. There are stronger options in this category.

Is GPT-5.3-Codex-Spark good for multilingual tasks?

GPT-5.3-Codex-Spark ranks #11 out of 123 models in multilingual tasks benchmarks with an average score of 90.8. There are stronger options in this category.

Which sibling models are related to GPT-5.3-Codex-Spark?

GPT-5.3-Codex-Spark belongs to the GPT-5.3 Codex family. Related variants on BenchLM include GPT-5.3 Codex.

What is the context window size of GPT-5.3-Codex-Spark?

GPT-5.3-Codex-Spark has a context window of 256K, which determines how much text it can process in a single interaction.

Last updated: March 12, 2026

Weekly LLM Updates

New model releases, benchmark scores, and leaderboard changes. Every Friday.

Free. Your signup is stored with a derived country code for compliance routing.