MODEL DIRECTORY

Follow the model.
See the evidence.

Benchmark results, evaluation settings and disclosed training methods—collected in one place for each model.

3,335 reported models and checkpoint variants. Scores keep their original units and sources.

546 models · page 6 of 12

OpenAI

GPT-3.5-turbo + EthicsPrompt + NoGoals

1 benchmarks · 4 observations

Explore results →

OpenAI

GPT-3.5-turbo-16K

1 benchmarks · 4 observations

Explore results →

OpenAI

GPT-3.5-Turbo-20231106†

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 (8-shot)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 (gpt-4-0613)

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4 (Medprompt) Microsoft (Nori et al. 2023)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 (N/A)

1 benchmarks · 4 observations

Explore results →

OpenAI

GPT-4 (NA)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 (vision preview)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Base

1 benchmarks · 3 observations

Explore results →

OpenAI

GPT-4 / Episodic + Reflexion

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Episodic Retrieval

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Reflexion

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Semantic + Episodic

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Semantic + Episodic + Reflexion

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Semantic + Reflexion

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Semantic Retrieval

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 / Zero-shot

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 + CoT

1 benchmarks · 4 observations

Explore results →

OpenAI

GPT-4 + CoT+ EthicsPrompt

1 benchmarks · 4 observations

Explore results →

OpenAI

GPT-4 + EthicsPrompt + NoGoals

1 benchmarks · 4 observations

Explore results →

OpenAI

GPT-4 + FreshPrompt

1 benchmarks · 4 observations

Explore results →

OpenAI

GPT-4 + plugins*

1 benchmarks · 3 observations

Explore results →

OpenAI

GPT-4 + Self-Ask

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 + Self-Ask (N/A)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 Baseline

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 Microsoft & OpenAI (Nori et al. 2023)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 Turbo 04/09

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4 w/ snippets only & random order

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4 w/ snippets only & search order

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4 w/ snippets only & time order

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4 with search

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-0314

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-0314 / human informal proofs

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4-0314 / model-generated informal proofs

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4-1106-preview-128k

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-base Microsoft & OpenAI (Nori et al. 2023)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-Preview-1106

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4-Turbo (0125) [37]

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-Turbo-0125

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-turbo-2024-04-09 (diff)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-Turbo-2024-04-09 (OpenAI, 2023)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4-turbo-2024-04-09 (udiff)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4.1 (Terminus 2)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4.5 Preview (February 2025)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4o (August 2024)

1 benchmarks · 1 observations

Explore results →

OpenAI

GPT-4o (Chinese)

1 benchmarks · 2 observations

Explore results →

OpenAI

GPT-4o (English subset)

1 benchmarks · 2 observations

Explore results →