Benchmark Almanac

OpenAI Research Engineer Interview: Multiple Choice

o4-mini 83% leads since April 2025
Only in system cards

Models answer 97 multiple-choice questions drawn from OpenAI machine learning interview topics. Scored as cons@32.

In-house OpenAI evaluation from the internal interview question bank.

Made by
OpenAI
Released
August 2024
Kind
Labs' internal R&D tests
Level
Expert
Scored
Checked automatically
Answers
Multiple choice
Tasks
Kept private
Strongest model tested
o3 #83 on the capability index
Last in a system card
July 2025
Last model added
July 2025
9
Models scored in lab cards
10
System cards report it, from 1 lab

o4-mini leads with 83% among models released by April 2025

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

The newest model with a score here came out in April 2025, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 83.0 o4-mini
  2. 80.0 o3-mini
  3. 80.0 GPT-4.5
  4. 80.0 o3
  5. 78.0 o1
  6. 78.0 OpenAI deep research
  7. 77.3 o1-mini
  8. 76.8 o1-preview
  9. 61.0 GPT-4o
  1. 1 o4-mini 83%
  2. 2 o3-mini 80%
  3. 3 GPT-4.5 80%
  4. 4 o3 80%
  5. 5 o1 78%
  6. 6 OpenAI deep research 78%
  7. 7 o1-mini 77.3%
  8. 8 o1-preview 76.8%
  9. 9 GPT-4o May 2024 61%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 o4-mini 83% Apr 2025 Lab model card
2 o3-mini 80% Jan 2025 Lab model card
3 GPT-4.5 80% Feb 2025 Lab model card
4 o3 80% Apr 2025 Lab model card
5 o1 78% Dec 2024 Lab model card
6 OpenAI deep research 78% Feb 2025 Lab model card
7 o1-mini 77.3% Sep 2024 Lab model card
8 o1-preview 76.8% Sep 2024 Lab model card
9 GPT-4o May 2024 61% Aug 2024 Lab model card

5 models have held number one in labs’ own numbers since its release in August 2024

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI
Lab claims Not tested since Apr 2025
o4-mini 83%

Source: Labs’ model cards

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) o4-mini o3-mini o1-mini GPT-4o 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) o4-mini o1-mini

Source: Labs’ model cards

Labs report it only in system cards

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.

10 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Jul 17, 2025 ChatGPT agent system card OpenAI RE Interview Multiple-Choice, cons@32, chart shows only ChatGPT Agent no-browsing helpful-only at 75%, no released-model value (Figure 14, 5.1.3.1) AI R&D risk
Apr 16, 2025 o4-mini system card 83 OpenAI RE Interview Multiple-Choice, cons@32, o4-mini no browsing launch candidate, printed chart label (Figure 16) AI R&D risk
Apr 16, 2025 o3 system card 80 OpenAI RE Interview Multiple-Choice, cons@32, o3 no browsing launch candidate, printed chart label (Figure 16) AI R&D risk
Feb 27, 2025 GPT-4.5 system card 80 OpenAI Research Engineer Interview, multiple choice, pre- and post-mitigation (4.6.1) Autonomy
Feb 25, 2025 OpenAI deep research system card 78 OpenAI RE Interview Multiple-Choice, cons@32, post-mitigation with browsing, text and printed chart label (without browsing 76%) (3.4.7) Autonomy
Jan 31, 2025 o3-mini system card 80 OpenAI RE Interview Multiple-Choice, cons@32, o3-mini (Post), printed chart label (5.7.1) Autonomy
Dec 5, 2024 o1 system card 78 OpenAI RE Interview Multiple-Choice, cons@32, o1 (post-mitigation), printed chart label (5.8.1) Autonomy
Sep 12, 2024 o1-preview system card 76.8 OpenAI Research Engineer Interviews (Multiple Choice), pass rate cons@32, o1-preview (post-mitigation), printed chart label (4.5.1) Autonomy
Sep 12, 2024 o1-mini system card 77.3 OpenAI Research Engineer Interviews (Multiple Choice), pass rate cons@32, o1-mini (post-mitigation), printed chart label (4.5.1) Autonomy
Aug 8, 2024 GPT-4o system card 61 OpenAI interview, multiple choice questions, cons@32 (3.8) AI R&D risk

More in AI research

All 104 in AI research
  • PaperBench In use Agent replicates ICML 2024 papers from scratch. LLM judge grades against author-written hierarchical rubrics. Qwen 3.8 Max 93%
  • MLS-Bench In use Agent must invent one algorithmic change in an ML research scaffold whose gain transfers across settings. Scored relative to human baselines. Claude Fable 5.1 50.3%
  • PostTrainBench 1.2 In use CLI agent gets a base model, one H100 and a time budget to post-train it for a target benchmark. Scored on normalized gain. Claude Fable 5.1 44.6%
  • MLE-bench In use Kaggle-style ML engineering competitions. Agent trains models and is scored against competition leaderboards. GPT-6 Astra 93.8%
  • KernelBench In use Replace PyTorch operators with faster custom GPU kernels. Scored on correctness and speedup. MiniMax-M3 28.8%
  • PMPP-Hard In use Agents write GPU kernels for 69 tasks that must be correct, and on 48 of them also run within 1.25 times the reference time. The score is the area under the solved-tasks curve over a 10-hour budget, out of 690 task-hours. Inkling 2 Small 494.3