Benchmark Almanac

SpaceXAI MTS Eval

Grok 4.6 61.1% Only in system cards

Agents do model-development tasks from xAI engineers, such as diagnosing reward hacking and debugging training infrastructure, under fixed budgets.

In-house xAI evaluation with 29 tasks in the standard non-GPU suite.

Made by xAI (SpaceXAI) Released Jul 2026 Labs' internal R&D tests Tasks kept private Last in a system card Aug 2026 Last model added Aug 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
1
Model scored in lab cards
2
System cards report it, from 1 lab

Grok 4.6 leads with 61.1%

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.

  1. 61.1 Grok 4.6
  1. 1 Grok 4.6 61.1%
#ModelScoreSettingDateSource
1 Grok 4.6 61.1% Aug 2026 Model card

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
0% 20% 40% 60% 80% 100% Near the ceiling (85%) Grok 4.6 0% 20% 40% 60% 80% 100% Near the ceiling (85%) Grok 4.6

Labs report it only in system cards

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark. Some system cards do, below.

2 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
12 Aug 2026 Grok 4.6 model card 61.1 SpaceXAI MTS Eval, standard non-GPU suite, % correct, high (5.1) AI R&D risk
14 Jul 2026 Grok 4.5 model card SpaceXAI MTS Eval, % correct vs mean tokens scatter, no printed value for Grok 4.5 (5.1) AI R&D risk

More in AI research

All 83 in AI research
  • 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%
  • PaperBench In use Agent replicates ICML 2024 papers from scratch. LLM judge grades against author-written hierarchical rubrics. Qwen 3.8 Max 93%
  • MLE-bench In use Kaggle-style ML engineering competitions. Agent trains models and is scored against competition leaderboards. GPT-6 Astra 93.8%
  • 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%
  • KernelBench In use Replace PyTorch operators with faster custom GPU kernels. Scored on correctness and speedup. MiniMax-M3 28.8%
  • OpenAI PRs Only in system cards Agents replicate internal OpenAI pull requests from a pre-PR branch and a prompt. Graded by hidden unit tests as pass@1. GPT-5.2 55%