#106 GPT-5.4

OpenAI · gpt-5.4

Released: 05.03.2026 · Source

GPT-5.4 is a model from OpenAI. Exact version: gpt-5.4. See the source for capabilities and limitations.

Use cases: Coding, Image understanding, Reasoning, Text
Accepts: Text & code, Images
Produces: Text & code
Context: 1050000 tokens

Access

OpenAI · API Computing: Cloud

Price USD

No verified data

Evaluator & score

Epoch AI · ECI156.86

Composite index · published model-group index · 20.09.2026 · Higher is better

Snapshot: 2026-09-20 · 90% CI [155.21, 159]

Official Epoch AI index; combines own, third-party and developer-reported results. It is not a safety audit or a media-generation score.

Conditions

Epoch Capabilities Index; official model-group estimates

Official ECI model scores, Epoch AI, CC BY 4.0. Composite of own, third-party and developer-reported results. 90% bootstrap interval. Not a per-effort score; export date is the model release date, not measurement date. Methodology: https://epoch.ai/eci#documentation

Epoch AI · Chess Puzzles38 %

Independent run · high · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Chess Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Chess Puzzles; version: gpt-5.4-2026-03-05_high; mode: high; source id: fQJLivo4QNbmQqnhyoXLav; file: chess_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Chess Puzzles20 %

Independent run · low · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Chess Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Chess Puzzles; version: gpt-5.4-2026-03-05_low; mode: low; source id: 5GLQdqyuDzgBmeZffzuwQx; file: chess_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Chess Puzzles38 %

Independent run · medium · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Chess Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Chess Puzzles; version: gpt-5.4-2026-03-05_medium; mode: medium; source id: n9rnjVzibawgtMunvu5yEp; file: chess_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Chess Puzzles5 %

Independent run · none · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Chess Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Chess Puzzles; version: gpt-5.4-2026-03-05_none; mode: none; source id: 6M8MmmtFv82WVQzUasKhYT; file: chess_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Chess Puzzles44 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Chess Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Chess Puzzles; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: RjvLecpyS4VVkJPSXhk8Ju; file: chess_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 11.03.2026

Epoch AI · EBR-bench25.397 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

EBR-bench; Epoch AI own run; best score across scorers

Epoch AI own run; EBR-bench; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: RNXPxRZ4vMiwbspiqETy8B; file: ebr_bench.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 25.06.2026

Epoch AI · FrontierMath47.6 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

FrontierMath; Epoch AI own run; best score across scorers

Epoch AI own run; FrontierMath-2025-02-28-Private; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: pre-release-gpt-5-4-xhigh-fm-t13; file: frontiermath.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 06.03.2026

Epoch AI · FrontierMath-Tier-4-2025-07-01-Private27.1 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

FrontierMath-Tier-4-2025-07-01-Private; Epoch AI own run; best score across scorers

Epoch AI own run; FrontierMath-Tier-4-2025-07-01-Private; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: pre-release-gpt-5-4-xhigh-fm-t4; file: frontiermath_tier_4.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 06.03.2026

Epoch AI · FrontierMath-Tier-4-v2-Private49 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

FrontierMath-Tier-4-v2-Private; Epoch AI own run; best score across scorers

Epoch AI own run; FrontierMath-Tier-4-v2-Private; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: doiQDghqnLzbE8CFijLNQJ; file: frontiermath_tier_4_v2.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 11.06.2026

Epoch AI · FrontierMath-Tiers-1-3-v2-Private78.596 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

FrontierMath-Tiers-1-3-v2-Private; Epoch AI own run; best score across scorers

Epoch AI own run; FrontierMath-Tiers-1-3-v2-Private; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: PaYEQ37A2n6giYBwZGBnXY; file: frontiermath_tiers_1_3_v2.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 11.06.2026

Epoch AI · GPQA Diamond89.899 %

Independent run · high · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

GPQA Diamond; Epoch AI own run; best score across scorers

Epoch AI own run; GPQA diamond; version: gpt-5.4-2026-03-05_high; mode: high; source id: FbhcT8NnJUntcyNGubJXY6; file: gpqa_diamond.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · GPQA Diamond84.848 %

Independent run · low · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

GPQA Diamond; Epoch AI own run; best score across scorers

Epoch AI own run; GPQA diamond; version: gpt-5.4-2026-03-05_low; mode: low; source id: aYgDc2iSDpmHkQfu7YjCHc; file: gpqa_diamond.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · GPQA Diamond88.889 %

Independent run · medium · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

GPQA Diamond; Epoch AI own run; best score across scorers

Epoch AI own run; GPQA diamond; version: gpt-5.4-2026-03-05_medium; mode: medium; source id: ManJk43Kf9VuXowNqoDrov; file: gpqa_diamond.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · GPQA Diamond74.747 %

Independent run · none · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

GPQA Diamond; Epoch AI own run; best score across scorers

Epoch AI own run; GPQA diamond; version: gpt-5.4-2026-03-05_none; mode: none; source id: LaBuzZExRqocD4FET6yx26; file: gpqa_diamond.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · GPQA Diamond93.3 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

GPQA Diamond; Epoch AI own run; best score across scorers

Epoch AI own run; GPQA diamond; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: fFatyce8UvpN7ZivdmrhAy; file: gpqa_diamond.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 06.03.2026

Epoch AI · MirrorCode15.556 %

Independent run · high · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

MirrorCode; Epoch AI own run; best score across scorers

Epoch AI own run; MirrorCode; version: gpt-5.4-2026-03-05_high; mode: high; source id: mirrorcode-ml-2l-gpt-5-4; file: mirrorcode.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 10.08.2026

Epoch AI · Mock AIME 2024/202597.778 %

Independent run · high · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mock AIME 2024/2025; Epoch AI own run; best score across scorers

Epoch AI own run; OTIS Mock AIME 2024-2025; version: gpt-5.4-2026-03-05_high; mode: high; source id: DpPahiEawokzgidE3cNZbS; file: otis_mock_aime_2024_2025.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Mock AIME 2024/202584.444 %

Independent run · low · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mock AIME 2024/2025; Epoch AI own run; best score across scorers

Epoch AI own run; OTIS Mock AIME 2024-2025; version: gpt-5.4-2026-03-05_low; mode: low; source id: bezRyWjXxqHnQeiL4MkSfZ; file: otis_mock_aime_2024_2025.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Mock AIME 2024/202595.556 %

Independent run · medium · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mock AIME 2024/2025; Epoch AI own run; best score across scorers

Epoch AI own run; OTIS Mock AIME 2024-2025; version: gpt-5.4-2026-03-05_medium; mode: medium; source id: eUJNMJSdLnJWNgRY8D2rYw; file: otis_mock_aime_2024_2025.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Mock AIME 2024/202557.778 %

Independent run · none · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mock AIME 2024/2025; Epoch AI own run; best score across scorers

Epoch AI own run; OTIS Mock AIME 2024-2025; version: gpt-5.4-2026-03-05_none; mode: none; source id: fjxs2EWts9GjpuBtFD5BN3; file: otis_mock_aime_2024_2025.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 15.07.2026

Epoch AI · Mock AIME 2024/202595.3 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mock AIME 2024/2025; Epoch AI own run; best score across scorers

Epoch AI own run; OTIS Mock AIME 2024-2025; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: HVsPchr34pWYHNuv5sKYyK; file: otis_mock_aime_2024_2025.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 06.03.2026

Epoch AI · Mystery Game Puzzles17 %

Independent run · low · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mystery Game Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Mystery Game Puzzles; version: gpt-5.4-2026-03-05_low; mode: low; source id: kWUUr3aHFZfKNARqAFxYZa; file: mystery_game_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 27.08.2026

Epoch AI · Mystery Game Puzzles28 %

Independent run · medium · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mystery Game Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Mystery Game Puzzles; version: gpt-5.4-2026-03-05_medium; mode: medium; source id: ereDhnap4G3Q6ydsNQW97W; file: mystery_game_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 28.08.2026

Epoch AI · Mystery Game Puzzles16 %

Independent run · none · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mystery Game Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Mystery Game Puzzles; version: gpt-5.4-2026-03-05_none; mode: none; source id: GtW2cWyAXZPCszLi63e4yJ; file: mystery_game_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 30.08.2026

Epoch AI · Mystery Game Puzzles37 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

Mystery Game Puzzles; Epoch AI own run; best score across scorers

Epoch AI own run; Mystery Game Puzzles; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: T6DyawEuhVmDAaDTzDRNNx; file: mystery_game_puzzles.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 24.07.2026

Epoch AI · SWE-bench Verified76.86 %

Independent run · high · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

SWE-bench Verified; Epoch AI own run; best score across scorers

Epoch AI own run; SWE-Bench verified; version: gpt-5.4-2026-03-05_high; mode: high; source id: i9HTMbAxcwsrTf2w2FuJ6N; file: swe_bench_verified.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 06.03.2026

Epoch AI · SimpleQA Verified45.1 %

Independent run · xhigh · 20.09.2026 · Higher is better

Snapshot: 2026-09-20

Conditions

SimpleQA Verified; Epoch AI own run; best score across scorers

Epoch AI own run; SimpleQA Verified; version: gpt-5.4-2026-03-05_xhigh; mode: xhigh; source id: AzCkm282PLhDdLJ8DnAQQc; file: simpleqa_verified.csv. CC BY 4.0, Epoch AI. Snapshot 2026-09-20. Source column Best score (across scorers), not AIpedia's selection of the best effort.

Measurement date: 27.08.2026

SourceVerified 20.09.2026