RankingEffort benchmarks

GPT-5.2 · XHigh effort benchmarks

Compare measured Low, Medium, High and other settings. Every number keeps its source.

Model

GPT-5.2

74 measured results · 5 reported effort labels · 43 benchmark/harness combinations. Model evidence and sources →

Capability score evidence

GPT-5.2 · XHigh: 31.2 overall comparison score. Meets the overall evidence requirements.

Which sources carry the weight?

Artificial Analysis · 37.4% direct comparison weight · independent
ARC Prize · 34.1% direct comparison weight · independent
SWE-rebench · 28.5% direct comparison weight · independent

These shares describe the comparisons entering the fit, not fractions of the final score.

How sensitive is this score?

Remove one evaluator or family and refit the graph against the same overall panel. Values use the matching dataset audited 2026-09-12. These are sensitivity checks, not confidence intervals; ranking eligibility may also change.

Removed evidence	Recomputed score	Change
Artificial Analysis	28.1	-3.0
ARC Prize	36.7	+5.5
Datacurve	27.8	-3.4
LiveCodeBench	31.2	+0.0
Scale	32.8	+1.6
Terminal-Bench	31.1	-0.0
Cognition	32.5	+1.3
LiveBench	30.9	-0.3
SWE-rebench	30.5	-0.6
Epoch AI	31.3	+0.1
Vals AI	30.3	-0.8
Provider reports · Anthropic	31.2	+0.0
Provider reports · Moonshot AI	31.1	-0.0
Provider reports · Meta	31.1	-0.0
Provider reports · Mistral	31.2	+0.0
All provider reports	31.1	-0.1

Every benchmark family

Removed evidence	Recomputed score	Change
arc-agi	36.5	+5.3
deepswe	27.8	-3.4
frontiercode	32.5	+1.3
mmmu-pro	32.3	+1.1
omniscience	30.5	-0.7
mmlu-pro	31.9	+0.7
simpleqa	31.8	+0.7
swe-rebench	30.5	-0.6
tau-bench	31.8	+0.6
hle	31.8	+0.6
aa-lcr	30.6	-0.5
epoch-game-puzzles	30.7	-0.5
ifbench	30.7	-0.4
gpqa	30.8	-0.4
scicode	31.5	+0.3
livebench-instructions	31.5	+0.3
multi-swe-bench	30.9	-0.3
swe-atlas-qna	31.0	-0.2
frontiermath	31.4	+0.2
harvey	31.0	-0.2
livebench-language	31.3	+0.2
apex-agents	31.3	+0.2
critpt	31.0	-0.2
livebench-coding	31.0	-0.1
vibe-code	31.0	-0.1
livebench-reasoning	31.0	-0.1
gdp-pdf	31.3	+0.1
livecodebench	31.3	+0.1
swe-atlas-test-writing	31.1	-0.1
briefcase	31.3	+0.1
vals-finance-agent	31.1	-0.1
gdpval	31.2	+0.1
vals-legal-research	31.1	-0.1
vals-code-migration	31.1	-0.1
aime	31.1	-0.1
swe-atlas-refactoring	31.1	-0.0
analyst-agent	31.2	+0.0
terminal-bench	31.2	+0.0
enterprise-ops	31.2	+0.0
vals-excel-modeling	31.1	-0.0
automationbench	31.2	+0.0
arc-agi-3	31.2	+0.0
livebench-data	31.2	+0.0
livebench-math	31.2	+0.0
itbench	31.2	+0.0
math500	31.2	-0.0
gmmlu	31.2	-0.0
enigma-eval	31.2	-0.0
swe-bench-pro	31.2	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 77 observations, 71 contributing, 0 matched without graph weight and 6 excluded. Counts do not establish rank eligibility.

3 · Benchmark family or source protocol has not been reviewed
3 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately

Inspect every observation and exclusion →

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AGPT-5.2 · XHigh3 of 7 capabilities supported

Gaps are unknown, not zero. Hollow points are preliminary.

Agentic
A17.3Preliminary
Hard reasoning
A35.8Supported
Coding
A32.9Supported
Human pref
Unknown
Knowledge
A36.8Supported
Multimodal
Unknown
Long context
A66.4Preliminary

Each profile belongs to the named configuration. Capability scores use a shared panel for all axes, distinct from the overall ranking panel; they are not accuracy percentages or direct head-to-head wins.

Compare this configuration →Explore the model evidence profile →

These are published benchmark scores. Supported effort estimates also appear on the Capability leaderboard. Results from different harnesses occupy different rows. Missing cells stay unknown; no scores are copied between effort levels.

Benchmark / source	Low	Medium	High	XHigh	Non-reasoning
AIME 2025Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	96.67% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	99% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	51% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
AA-LCR v1.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	70.33% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	82.67% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	45.67% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
CritPtArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by Argonne and UIUC, with contributions from 60+ researchers globally · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	7.86% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	11.55% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	0.57% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
Global-MMLU-LiteArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	89.76% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	Not reported	86.8% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
GPQA DiamondArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	86.36% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	90.3% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	71.21% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
Humanity's Last ExamArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	26.69% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	37.67% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	8.02% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
IFBenchArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	65.24% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	75.44% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	47.41% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
LiveCodeBenchArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	89.42% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	88.89% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	66.88% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
MMLU-ProArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	85.85% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	87.36% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	81.36% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
MMMU-ProArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	74.57% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	Not reported	65.84% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
AA-Omniscience IndexArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis Unit: index. Source collected 2026-09-12.	Not reported	0.27 Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	-0.88 Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	-12.25 Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
AA-Omniscience AccuracyArtificial Analysis · 6000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader Benchmark details AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	38.28% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	44.32% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	30.85% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
𝜏²-Bench TelecomArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	74.27% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	84.8% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	46.49% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
𝜏³-BankingArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by Sierra Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	Not reported	Not reported	11.13% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
Terminal-Bench HardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	43.18% Reported settings GPT-5.2 (medium); Model source: https://artificialanalysis.ai/models/gpt-5-2-medium	Not reported	46.97% Reported settings GPT-5.2 (xhigh); Model source: https://artificialanalysis.ai/models/gpt-5-2	31.82% Reported settings GPT-5.2 (Non-reasoning); Model source: https://artificialanalysis.ai/models/gpt-5-2-non-reasoning
ARC-AGI-2ARC Prize · v2 Semi-Private Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-09-12.	9.72% Reported settings GPT-5.2 (Low); Source model ID: gpt-5-2-2025-12-11-thinking-low; 120 tasks	26.67% Reported settings GPT-5.2 (Medium); Source model ID: gpt-5-2-2025-12-11-thinking-medium; 120 tasks	43.33% Reported settings GPT-5.2 (High); Source model ID: gpt-5-2-2025-12-11-thinking-high; 120 tasks	52.91% Reported settings GPT-5.2 (XHigh); Source model ID: gpt-5-2-2025-12-11-thinking-xhigh; 120 tasks	Not reported
LiveBench · AMPS Hard · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	97% Reported settings LiveBench official task evaluation; livebench-2026-06-25-AMPS_Hard; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · code completion · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	76.09% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_completion; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · code generation · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	76.06% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_generation; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · connections · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	99% Reported settings LiveBench official task evaluation; livebench-2026-06-25-connections; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · consecutive events · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	88.91% Reported settings LiveBench official task evaluation; livebench-2026-06-25-consecutive_events; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · integrals with game · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	92% Reported settings LiveBench official task evaluation; livebench-2026-06-25-integrals_with_game; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · javascript · 2026-06-25LiveBench · LiveBench Mini-SWE-Agent, 250-step limit Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	59.09% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-javascript; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · logic with navigation · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	74% Reported settings LiveBench official task evaluation; livebench-2026-06-25-logic_with_navigation; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · math comp · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	95.1% Reported settings LiveBench official task evaluation; livebench-2026-06-25-math_comp; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · olympiad · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	88.57% Reported settings LiveBench official task evaluation; livebench-2026-06-25-olympiad; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · paraphrase · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	64.1% Reported settings LiveBench official task evaluation; livebench-2026-06-25-paraphrase; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · plot unscrambling · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	58.43% Reported settings LiveBench official task evaluation; livebench-2026-06-25-plot_unscrambling; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · python · 2026-06-25LiveBench · LiveBench Mini-SWE-Agent, 250-step limit Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	55% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-python; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · simplify · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	54.08% Reported settings LiveBench official task evaluation; livebench-2026-06-25-simplify; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · spatial · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	96% Reported settings LiveBench official task evaluation; livebench-2026-06-25-spatial; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · story generation · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	65.7% Reported settings LiveBench official task evaluation; livebench-2026-06-25-story_generation; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · summarize · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	63.2% Reported settings LiveBench official task evaluation; livebench-2026-06-25-summarize; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · tablejoin · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	45.58% Reported settings LiveBench official task evaluation; livebench-2026-06-25-tablejoin; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · tablereformat · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	100% Reported settings LiveBench official task evaluation; livebench-2026-06-25-tablereformat; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · theory of mind · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	78.85% Reported settings LiveBench official task evaluation; livebench-2026-06-25-theory_of_mind; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · typescript · 2026-06-25LiveBench · LiveBench Mini-SWE-Agent, 250-step limit Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	36.67% Reported settings LiveBench Mini-SWE-Agent, 250-step limit; livebench-2026-06-25-typescript; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · typos · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	82% Reported settings LiveBench official task evaluation; livebench-2026-06-25-typos; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
LiveBench · zebra puzzle · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	84% Reported settings LiveBench official task evaluation; livebench-2026-06-25-zebra_puzzle; gpt-5.2-2025-12-11-high; checkpoint=gpt-5.2	Not reported	Not reported
SWE-rebench · 2025-11 · toolsSWE-rebench · SWE-rebench standardized ReAct tools Benchmark details Average resolved rate across repeated runs, not best-of-N success. Disjoint complete calendar months within each model’s tested range; standardized harness and all-language task set. Unit: percent. Source collected 2026-09-12.	Not reported	61.28% Reported settings SWE-rebench standardized ReAct tools; swe-rebench-2025-11-tools; gpt-5.2-2025-12-11-medium; checkpoint=gpt-5.2	Not reported	Not reported	Not reported
SWE-rebench · 2025-12 · toolsSWE-rebench · SWE-rebench standardized ReAct tools Benchmark details Average resolved rate across repeated runs, not best-of-N success. Disjoint complete calendar months within each model’s tested range; standardized harness and all-language task set. Unit: percent. Source collected 2026-09-12.	Not reported	59.41% Reported settings SWE-rebench standardized ReAct tools; swe-rebench-2025-12-tools; gpt-5.2-2025-12-11-medium; checkpoint=gpt-5.2	Not reported	61.49% Reported settings SWE-rebench standardized ReAct tools; swe-rebench-2025-12-tools; gpt-5.2-2025-12-11-xhigh; checkpoint=gpt-5.2	Not reported
SWE-rebench · 2026-01 · toolsSWE-rebench · SWE-rebench standardized ReAct tools Benchmark details Average resolved rate across repeated runs, not best-of-N success. Disjoint complete calendar months within each model’s tested range; standardized harness and all-language task set. Unit: percent. Source collected 2026-09-12.	Not reported	51.05% Reported settings SWE-rebench standardized ReAct tools; swe-rebench-2026-01-tools; gpt-5.2-2025-12-11-medium; checkpoint=gpt-5.2	Not reported	51.67% Reported settings SWE-rebench standardized ReAct tools; swe-rebench-2026-01-tools; gpt-5.2-2025-12-11-xhigh; checkpoint=gpt-5.2	Not reported
SWE-rebench · 2026-02 · toolsSWE-rebench · SWE-rebench standardized ReAct tools Benchmark details Average resolved rate across repeated runs, not best-of-N success. Disjoint complete calendar months within each model’s tested range; standardized harness and all-language task set. Unit: percent. Source collected 2026-09-12.	Not reported	64.42% Reported settings SWE-rebench standardized ReAct tools; swe-rebench-2026-02-tools; gpt-5.2-2025-12-11-medium; checkpoint=gpt-5.2	Not reported	Not reported	Not reported

What these comparisons mean

Each source reports its own effort labels. Matching labels do not establish equal compute budgets or identical fallback behavior. Open Reported settings to inspect the exact model name, agent and task coverage. Multiple results at one level remain visible rather than selecting the best. Source-reported None or Non-reasoning labels are not treated as evidence that an API supports those settings.

These measurements use their own dated source collection. The leaderboard fits identified configurations jointly; it contains no pooled model ratings. Explicit thinking budgets are separate configurations. Read the methodology.

Download all effort benchmark results · 7,424 results across 240 models · Collected 2026-09-12

GPT-5.2 · XHigh effort benchmarks

GPT-5.2

Capability score evidence

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context