Language modelAPI only

GPT-5.3-Codex

OpenAI

Released: February 5, 2026
Data date: September 8, 2026

Intelligence: Not ranked
Overall score
Speed: 139.4 tokens/s
Output throughput
Price: $1.75 · $14
Input · Output / 1M tokens
Input: Text, Image
Output: Text

No published category-wide ranking is available for this model.

Entries 1-12 of 31. Page 1 of 3.

Published benchmarks

Each chart compares models in the same documented test and highlights the current model.

Agent tasks

2 measurements

METR Time Horizon

50% time horizon

What this benchmark measures: Human task duration at which a model-agent system reaches a 50% or 80% predicted success probability.

Context: Minutes refer to human task duration, not model runtime. METR considers estimates above 16 hours unreliable.

Higher is better4 modelsGPT-5.3-Codex: Rank 2 of 4Data date: August 26, 2026

1.Gemini 3.1 Pro Preview384.15 minutes

2.GPT-5.3-Codex349.53 minutes

3.GPT-5203.01 minutes

4.o3119.73 minutes

Scale: 0 minutes - 384.15 minutes

Source: METR

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: metr-th1-1-p50
Metric: 50% time horizon
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 4
Methodology and context: Triframe agent with six independent runs per task. Time is estimated human completion duration.. Source model ID: curated/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Model snapshot: gpt-5.3-codex-2026-02-05. Evaluator: METR. Dataset type: public. Leaderboard date: 2026-05-08. Harness: Triframe. Agent: Triframe. Attempts: 6 independent runs per task. Score display: 5.8 h (95% confidence interval 3.2 h to 13.6 h). Source ID: metr. Evidence capture ID: curated-metr-2026-08-26-eae972724332. Source payload hash: eae972724332b484893349fcc227a9b865d78efcd94e3b78d2d11cbc2315801e. Parser version: curated-v1. Imported at: 2026-09-04

Source: METR

METR Time Horizon

80% time horizon

What this benchmark measures: Human task duration at which a model-agent system reaches a 50% or 80% predicted success probability.

Context: Minutes refer to human task duration, not model runtime. METR considers estimates above 16 hours unreliable.

Higher is better4 modelsGPT-5.3-Codex: Rank 2 of 4Data date: August 26, 2026

1.Gemini 3.1 Pro Preview89.8 minutes

2.GPT-5.3-Codex54.74 minutes

3.GPT-538.31 minutes

4.o329.98 minutes

Scale: 0 minutes - 89.8 minutes

Source: METR

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: metr-th1-1-p80
Metric: 80% time horizon
Scale: No upper bound is documented. The chart spans the recorded values and zero.
Participants: 4
Methodology and context: Triframe agent with six independent runs per task. Time is estimated human completion duration.. Source model ID: curated/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Model snapshot: gpt-5.3-codex-2026-02-05. Evaluator: METR. Dataset type: public. Leaderboard date: 2026-05-08. Harness: Triframe. Agent: Triframe. Attempts: 6 independent runs per task. Score display: 54.7 min (95% confidence interval 22.4 min to 2 h). Source ID: metr. Evidence capture ID: curated-metr-2026-08-26-bdb68f137fa5. Source payload hash: bdb68f137fa51fbbd7af0345aacd4463b88a571dc43361afdd115bbe86d87dc2. Parser version: curated-v1. Imported at: 2026-09-04

Source: METR

Coding

10 measurements

IOI

IOI 2024

Higher is better4 modelsGPT-5.3-Codex: Rank 3 of 4Data date: August 11, 2026

1.GPT-5.467.17%

2.GPT-5.6 Terra66.17%

3.GPT-5.3-Codex58.17%

4.GPT-5.4 mini6.67%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-ioi:1:ioi2024
Metric: accuracy
Scale: 0-100
Participants: 4
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 2631.97. Standard error: 14.481. Cost per task in USD: 1.818037. Source ID: vals-ai. Evidence capture ID: vals-ioi-2026-08-11-2dfe8b4b8b7b. Source payload hash: 2dfe8b4b8b7be768b4617349894b78a429aa610031aff027b405096fb56d23cc. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

IOI

IOI 2025

Higher is better4 modelsGPT-5.3-Codex: Rank 3 of 4Data date: August 11, 2026

1.GPT-5.468.5%

2.GPT-5.6 Terra64.33%

3.GPT-5.3-Codex29.5%

4.GPT-5.4 mini6.17%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-ioi:1:ioi2025
Metric: accuracy
Scale: 0-100
Participants: 4
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 4516.544. Standard error: 5.854. Cost per task in USD: 2.222013. Source ID: vals-ai. Evidence capture ID: vals-ioi-2026-08-11-2dfe8b4b8b7b. Source payload hash: 2dfe8b4b8b7be768b4617349894b78a429aa610031aff027b405096fb56d23cc. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

IOI

Overall

Higher is better24 modelsGPT-5.3-Codex: Rank 7 of 24Data date: September 4, 2026

1.Claude Opus 591.67%

2.GPT-5.6 Sol86.67%

3.GPT-5.6 Luna72.92%

4.Claude Fable 572.25%

5.GPT-5.467.83%

6.GPT-5.6 Terra65.25%

7.GPT-5.3-Codex43.83%

8.Gemini 3 Flash39.08%

9.DeepSeek-V4-Pro35.83%

10.Gemini 3.5 Flash-Lite26.17%

11.GLM-522%

12.GPT-520%

13.Kimi K2.517.67%

14.Gemini 2.5 Pro17.08%

15.Grok 4.315.33%

16.GPT-5.4 nano15.25%

17.GLM-4.77.58%

18.GPT-5 mini6.75%

19.GPT-5.4 mini6.42%

20.o4-mini4.83%

21.GLM-4.64.33%

22.GLM-4.52.92%

23.Gemini 2.5 Flash2.61%

24.DeepSeek-V3 03241.67%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-ioi:1:overall
Metric: accuracy
Scale: 0-100
Participants: 24
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Latency in seconds: 3574.257. Cost per task in USD: 2.020025. Source ID: vals-ai. Evidence capture ID: vals-ioi-2026-09-04-c6f99e8d7130. Source payload hash: c6f99e8d713092dbfdc9cbcfd6b376fbb7209970714528eab31f6b541d90f20c. Parser version: vals-astro-v2. Imported at: 2026-09-04

Source: Vals AI

LiveCodeBench v6

Easy

Higher is better5 modelsGPT-5.3-Codex: Rank 1 of 5Data date: August 11, 2026

1.GPT-5.3-Codex99.07%

1.GPT-5.499.07%

3.GPT-5.598.45%

4.GPT-5.4 mini97.83%

5.GPT-5.6 Terra97.21%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-lcb:1:easy
Metric: accuracy
Scale: 0-100
Participants: 5
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 17.469. Standard error: 0.535. Cost per task in USD: 0.013703. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-08-11-0342bcc364e4. Source payload hash: 0342bcc364e4426ffb24589c2752c9225be61fbd26db8deba5bdab9775ce5a75. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

LiveCodeBench v6

Hard

Higher is better5 modelsGPT-5.3-Codex: Rank 1 of 5Data date: August 11, 2026

1.GPT-5.3-Codex75.14%

2.GPT-5.6 Terra72.86%

3.GPT-5.569.71%

4.GPT-5.467.71%

5.GPT-5.4 mini61.71%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-lcb:1:hard
Metric: accuracy
Scale: 0-100
Participants: 5
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 306.379. Standard error: 2.31. Cost per task in USD: 0.209533. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-08-11-0342bcc364e4. Source payload hash: 0342bcc364e4426ffb24589c2752c9225be61fbd26db8deba5bdab9775ce5a75. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

LiveCodeBench v6

Medium

Higher is better5 modelsGPT-5.3-Codex: Rank 1 of 5Data date: August 11, 2026

1.GPT-5.3-Codex87.73%

1.GPT-5.587.73%

1.GPT-5.6 Terra87.73%

4.GPT-5.485.64%

5.GPT-5.4 mini84.86%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-lcb:1:medium
Metric: accuracy
Scale: 0-100
Participants: 5
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 74.045. Standard error: 1.677. Cost per task in USD: 0.051518. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-08-11-0342bcc364e4. Source payload hash: 0342bcc364e4426ffb24589c2752c9225be61fbd26db8deba5bdab9775ce5a75. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

LiveCodeBench v6

Overall

Higher is better56 modelsGPT-5.3-Codex: Rank 11 of 56Data date: September 4, 2026

1.Claude Fable 589.78%

2.Gemini 3.8 Flash89.48%

3.Claude Opus 589.03%

4.Gemini 3.7 Flash88.65%

5.Gemini 3.1 Pro Preview88.49%

6.Grok 4.688.22%

7.Gemini 3.6 Flash88.08%

8.Gemini 3.5 Flash87.6%

9.DeepSeek-V4-Pro87.48%

10.Grok 4.587.35%

11.GPT-5.3-Codex87.31%

12.Kimi K387.19%

13.Kimi K2.686.77%

14.GPT-5 mini86.61%

15.Inkling Small85.93%

16.GPT-5.6 Terra85.93%

17.GPT-585.91%

18.Gemini 3 Flash85.59%

19.Inkling85.52%

20.GPT-5.585.3%

21.Grok 4.384.49%

22.GPT-5.484.14%

23.GPT-5.4 nano84.01%

24.o383.91%

25.Kimi K2.583.87%

26.GPT OSS 120B83.23%

27.GPT-5.6 Sol82.6%

28.Claude Sonnet 582.43%

29.GLM-4.782.23%

30.o4-mini82.21%

31.Kimi K2.7 Code82.05%

32.GLM-581.87%

33.MiMo-V2.581.51%

34.GPT-5.4 mini81.47%

35.GLM-5.181.38%

36.MiMo-V2.5-Pro81.35%

37.GLM-4.681.04%

38.GLM-5.380.54%

39.GLM-5.3-Flash80.51%

40.GPT OSS 20B80.39%

41.Gemini 3.1 Flash-Lite80.12%

42.Gemini 3.5 Flash-Lite79.01%

43.o3-mini71.48%

44.DeepSeek-R170.22%

45.GPT-5 nano70.22%

46.GLM-5.269.5%

47.GLM-4.567.45%

48.DeepSeek-V3 032465.48%

49.Kimi K2 Thinking63.15%

50.GPT-4.1 mini58.16%

51.GPT-4.154.67%

52.o150.26%

53.GPT-4o43.44%

54.GPT-4.1 nano42.72%

55.GPT-4o mini26.42%

56.Command R+18.24%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-lcb:1:overall
Metric: accuracy
Scale: 0-100
Participants: 56
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-09-01. Latency in seconds: 132.631. Cost per task in USD: 0.091585. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-09-04-95ca83387127. Source payload hash: 95ca83387127df287e9a6d928c411c98c9b65b4cefeee46eb4008ca1521f5ed4. Parser version: vals-astro-v2. Imported at: 2026-09-04

Source: Vals AI

SWE-bench Verified

<15 min fix

Higher is better5 modelsGPT-5.3-Codex: Rank 2 of 5Data date: August 11, 2026

1.GPT-5.591.75%

2.GPT-5.3-Codex90.21%

3.GPT-5.488.14%

4.GPT-5.4 mini85.57%

5.GPT-5.6 Terra77.84%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-swebench:1:<15 min fix
Metric: accuracy
Scale: 0-100
Participants: 5
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-08. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 185.93. Standard error: 2.14. Cost per task in USD: 0.340336. Source ID: vals-ai. Evidence capture ID: vals-swebench-2026-08-11-a6212cd857d8. Source payload hash: a6212cd857d8406828224bea2aee5b657d927338e035f1ff603eea74c9bc8a7b. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

SWE-bench Verified

>4 hours

Higher is better5 modelsGPT-5.3-Codex: Rank 2 of 5Data date: August 11, 2026

1.GPT-5.566.67%

2.GPT-5.3-Codex33.33%

2.GPT-5.6 Terra33.33%

4.GPT-5.40%

4.GPT-5.4 mini0%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-swebench:1:>4 hours
Metric: accuracy
Scale: 0-100
Participants: 5
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-08. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 412.097. Standard error: 33.333. Cost per task in USD: 0.69549. Source ID: vals-ai. Evidence capture ID: vals-swebench-2026-08-11-a6212cd857d8. Source payload hash: a6212cd857d8406828224bea2aee5b657d927338e035f1ff603eea74c9bc8a7b. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

SWE-bench Verified

1-4 hours

Higher is better5 modelsGPT-5.3-Codex: Rank 2 of 5Data date: August 11, 2026

1.GPT-5.6 Terra64.29%

2.GPT-5.3-Codex54.76%

3.GPT-5.450%

3.GPT-5.550%

5.GPT-5.4 mini33.33%

Scale: 0% - 100%

Source: Vals AI

CC BY 4.0

gradually.ai

Methodology and context

Comparison cohort: vals-swebench:1:1-4 hours
Metric: accuracy
Scale: 0-100
Participants: 5
Methodology and context: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-08. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 405.865. Standard error: 7.773. Cost per task in USD: 0.818585. Source ID: vals-ai. Evidence capture ID: vals-swebench-2026-08-11-a6212cd857d8. Source payload hash: a6212cd857d8406828224bea2aee5b657d927338e035f1ff603eea74c9bc8a7b. Parser version: vals-astro-v2. Imported at: 2026-08-11

Source: Vals AI

Recorded measurement details

Every observation retains its source value and published test conditions.

METR Time Horizon: 50% time horizon (Show measurement, test conditions, and source)

Source value: 349.530732
Score: 349.530732
Metric: 50% time horizon
Unit: minutes
Benchmark version: TH 1.1
Category: agentic
Task: p50
Task label: 50% time horizon
Direction: Higher is better
Comparison cohort: metr-th1-1-p50
Source type: independent-evaluator
Evaluator: METR
Status: active
Retrieved at: 2026-08-26
Methodology: Triframe agent with six independent runs per task. Time is estimated human completion duration.. Source model ID: curated/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Model snapshot: gpt-5.3-codex-2026-02-05. Evaluator: METR. Dataset type: public. Leaderboard date: 2026-05-08. Harness: Triframe. Agent: Triframe. Attempts: 6 independent runs per task. Score display: 5.8 h (95% confidence interval 3.2 h to 13.6 h). Source ID: metr. Evidence capture ID: curated-metr-2026-08-26-eae972724332. Source payload hash: eae972724332b484893349fcc227a9b865d78efcd94e3b78d2d11cbc2315801e. Parser version: curated-v1. Imported at: 2026-09-04
Comparison methodology: Triframe agent with six independent runs per task. Time is estimated human completion duration.

Source METR

METR Time Horizon: 80% time horizon (Show measurement, test conditions, and source)

Source value: 54.739407
Score: 54.739407
Metric: 80% time horizon
Unit: minutes
Benchmark version: TH 1.1
Category: agentic
Task: p80
Task label: 80% time horizon
Direction: Higher is better
Comparison cohort: metr-th1-1-p80
Source type: independent-evaluator
Evaluator: METR
Status: active
Retrieved at: 2026-08-26
Methodology: Triframe agent with six independent runs per task. Time is estimated human completion duration.. Source model ID: curated/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Model snapshot: gpt-5.3-codex-2026-02-05. Evaluator: METR. Dataset type: public. Leaderboard date: 2026-05-08. Harness: Triframe. Agent: Triframe. Attempts: 6 independent runs per task. Score display: 54.7 min (95% confidence interval 22.4 min to 2 h). Source ID: metr. Evidence capture ID: curated-metr-2026-08-26-bdb68f137fa5. Source payload hash: bdb68f137fa51fbbd7af0345aacd4463b88a571dc43361afdd115bbe86d87dc2. Parser version: curated-v1. Imported at: 2026-09-04
Comparison methodology: Triframe agent with six independent runs per task. Time is estimated human completion duration.

Source METR

IOI: IOI 2024 (Show measurement, test conditions, and source)

Source value: 58.167
Score: 58.167
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: ioi2024
Task label: IOI 2024
Direction: Higher is better
Comparison cohort: vals-ioi:1:ioi2024
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 2631.97. Standard error: 14.481. Cost per task in USD: 1.818037. Source ID: vals-ai. Evidence capture ID: vals-ioi-2026-08-11-2dfe8b4b8b7b. Source payload hash: 2dfe8b4b8b7be768b4617349894b78a429aa610031aff027b405096fb56d23cc. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

IOI: IOI 2025 (Show measurement, test conditions, and source)

Source value: 29.5
Score: 29.5
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: ioi2025
Task label: IOI 2025
Direction: Higher is better
Comparison cohort: vals-ioi:1:ioi2025
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 4516.544. Standard error: 5.854. Cost per task in USD: 2.222013. Source ID: vals-ai. Evidence capture ID: vals-ioi-2026-08-11-2dfe8b4b8b7b. Source payload hash: 2dfe8b4b8b7be768b4617349894b78a429aa610031aff027b405096fb56d23cc. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

IOI: Overall (Show measurement, test conditions, and source)

Source value: 43.834
Score: 43.834
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-ioi:1:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-09-04
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Latency in seconds: 3574.257. Cost per task in USD: 2.020025. Source ID: vals-ai. Evidence capture ID: vals-ioi-2026-09-04-c6f99e8d7130. Source payload hash: c6f99e8d713092dbfdc9cbcfd6b376fbb7209970714528eab31f6b541d90f20c. Parser version: vals-astro-v2. Imported at: 2026-09-04

Source Vals AI

LiveCodeBench v6: Easy (Show measurement, test conditions, and source)

Source value: 99.068
Score: 99.068
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: easy
Task label: Easy
Direction: Higher is better
Comparison cohort: vals-lcb:1:easy
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 17.469. Standard error: 0.535. Cost per task in USD: 0.013703. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-08-11-0342bcc364e4. Source payload hash: 0342bcc364e4426ffb24589c2752c9225be61fbd26db8deba5bdab9775ce5a75. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

LiveCodeBench v6: Hard (Show measurement, test conditions, and source)

Source value: 75.143
Score: 75.143
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: hard
Task label: Hard
Direction: Higher is better
Comparison cohort: vals-lcb:1:hard
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 306.379. Standard error: 2.31. Cost per task in USD: 0.209533. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-08-11-0342bcc364e4. Source payload hash: 0342bcc364e4426ffb24589c2752c9225be61fbd26db8deba5bdab9775ce5a75. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

LiveCodeBench v6: Medium (Show measurement, test conditions, and source)

Source value: 87.728
Score: 87.728
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: medium
Task label: Medium
Direction: Higher is better
Comparison cohort: vals-lcb:1:medium
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-09. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 74.045. Standard error: 1.677. Cost per task in USD: 0.051518. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-08-11-0342bcc364e4. Source payload hash: 0342bcc364e4426ffb24589c2752c9225be61fbd26db8deba5bdab9775ce5a75. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

LiveCodeBench v6: Overall (Show measurement, test conditions, and source)

Source value: 87.313
Score: 87.313
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-lcb:1:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-09-04
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-09-01. Latency in seconds: 132.631. Cost per task in USD: 0.091585. Source ID: vals-ai. Evidence capture ID: vals-lcb-2026-09-04-95ca83387127. Source payload hash: 95ca83387127df287e9a6d928c411c98c9b65b4cefeee46eb4008ca1521f5ed4. Parser version: vals-astro-v2. Imported at: 2026-09-04

Source Vals AI

SWE-bench Verified: <15 min fix (Show measurement, test conditions, and source)

Source value: 90.206
Score: 90.206
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: <15 min fix
Task label: <15 min fix
Direction: Higher is better
Comparison cohort: vals-swebench:1:<15 min fix
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-08. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 185.93. Standard error: 2.14. Cost per task in USD: 0.340336. Source ID: vals-ai. Evidence capture ID: vals-swebench-2026-08-11-a6212cd857d8. Source payload hash: a6212cd857d8406828224bea2aee5b657d927338e035f1ff603eea74c9bc8a7b. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

SWE-bench Verified: >4 hours (Show measurement, test conditions, and source)

Source value: 33.333
Score: 33.333
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: >4 hours
Task label: >4 hours
Direction: Higher is better
Comparison cohort: vals-swebench:1:>4 hours
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-08. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 412.097. Standard error: 33.333. Cost per task in USD: 0.69549. Source ID: vals-ai. Evidence capture ID: vals-swebench-2026-08-11-a6212cd857d8. Source payload hash: a6212cd857d8406828224bea2aee5b657d927338e035f1ff603eea74c9bc8a7b. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

SWE-bench Verified: 1-4 hours (Show measurement, test conditions, and source)

Source value: 54.762
Score: 54.762
Metric: accuracy
Unit: %
Benchmark version: 1
Category: coding
Task: 1-4 hours
Task label: 1-4 hours
Direction: Higher is better
Comparison cohort: vals-swebench:1:1-4 hours
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-08-11
Methodology: Source model ID: openai/gpt-5.3-codex. Source model name: GPT-5.3-Codex. Provider reported by source: OpenAI. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-08-08. Reasoning effort: xhigh. Maximum output tokens: 128000. Latency in seconds: 405.865. Standard error: 7.773. Cost per task in USD: 0.818585. Source ID: vals-ai. Evidence capture ID: vals-swebench-2026-08-11-a6212cd857d8. Source payload hash: a6212cd857d8406828224bea2aee5b657d927338e035f1ff603eea74c9bc8a7b. Parser version: vals-astro-v2. Imported at: 2026-08-11
Comparison methodology: Reasoning effort: xhigh. Maximum output tokens: 128000

Source Vals AI

Our model tests

Community tests you can inspect

Compare how the models respond to the same prompt. Each test shows the first attempt, with no subsequent fixes to the generated code. These results do not contribute to an overall score.

Pelican on a bicycle

A community classic for free-form SVG drawing.

GPT-5.3-Codex

Not tested yet

Prompt and test conditions

Generate an SVG of a pelican riding a bicycle

Token limit including reasoning: 8,192 tokens

Task origin (Simon Willison)

GPT-5.3-Codex

Published benchmarks

Agent tasks

METR Time Horizon

METR Time Horizon

Coding

IOI

IOI

IOI

LiveCodeBench v6

LiveCodeBench v6

LiveCodeBench v6

LiveCodeBench v6

SWE-bench Verified

SWE-bench Verified

SWE-bench Verified

Recorded measurement details

Community tests you can inspect

Pelican on a bicycle

Specifications and access

Additional source observations

Published prices

100 API requests

Compare this model

Sources and data date