GLM-5.2

Leaderboards

Where GLM-5.2 sits among the top 20 in each category.

By task

Scored from votes on each kind of prompt alone, on the same 1,000-point scale.

Test results

Every test GLM-5.2 has a published result on, out of 100%, with its rank among the models that took it.

• GPQA

  #22 / 256

  Graduate-level science questions written to be hard to look up.

  91.2%

  Self-reported

• Humanity's Last Exam

  #15 / 108

  Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

  54.7%

  Self-reported · In the text score· In the vision score

• SWE-Bench Pro

  #17 / 61

  Harder, longer software engineering tasks from real codebases.

  62.1%

  Self-reported · In the coding score

• DeepSWE 1.1

  #44 / 50

  DeepSWE 1.1 evaluates software engineering agents using the mini-swe-agent harness.

  44.0%

  Self-reported · In the coding score

• Terminal-Bench 2.1

  #20 / 46

  Terminal-Bench 2.1 is an updated release of the Terminal-Bench benchmark that tests AI agents' ability to operate a computer via the terminal. It evaluates how well models handle real-world, end-to-end tasks autonomously, including compiling code, training models, setting up servers, system administration, data science workflows, and security tasks.

  82.7%

  Self-reported

• Toolathlon

  #27 / 42

  Tool Decathlon is a comprehensive benchmark for evaluating AI agents' ability to use multiple tools across diverse task categories. It measures proficiency in tool selection, sequencing, and execution across ten different tool-use scenarios.

  48.2%

  Self-reported · In the agents score

• MCP Atlas

  #12 / 38

  MCP Atlas is a benchmark for evaluating AI models on scaled tool use capabilities, measuring how well models can coordinate and utilize multiple tools across complex multi-step tasks.

  76.8%

  Self-reported

• HMMT 2025

  #9 / 33

  Harvard-MIT Mathematics Tournament 2025 - A prestigious student-organized mathematics competition for high school students featuring two tournaments (November 2025 at MIT and February 2026 at Harvard) with individual tests, team rounds, and guts rounds

  94.4%

  Self-reported

• AIME 2026

  #1 / 26

  All 30 problems from the 2026 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.

  99.2%

  Self-reported

• NL2Repo

  #11 / 25

  NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire repositories from natural language specifications.

  48.9%

  Self-reported · In the coding score

• FrontierCode 1.1

  #20 / 23

  FrontierCode 1.1 evaluates whether coding-agent changes are mergeable, using unit tests, maintainer-defined rubrics, and verifiers. Runs flagged for unfair internet use receive a zero score.

  24.5%

  Self-reported · In the coding score

• IMO-AnswerBench

  #2 / 22

  IMO-AnswerBench is a benchmark for evaluating mathematical reasoning capabilities on International Mathematical Olympiad (IMO) problems, focusing on answer generation and verification.

  91.0%

  Self-reported

• Program Bench

  #5 / 18

  Program Bench evaluates code-generation agents by asking them to recreate a program's behavior from only a compiled binary and documentation. It spans 200 tasks from small CLI tools to large systems such as FFmpeg and SQLite, with submissions judged against more than 248,000 fuzz-generated behavioral tests.

  63.7%

  Self-reported · In the coding score

• FrontierSWE

  #5 / 16

  FrontierSWE measures whether an agent can complete open-ended technical projects at the scale of hours to tens of hours, spanning systems optimization, large-scale code construction, and applied ML research. Performance is reported as a dominance score, where higher is better.

  74.0%

  Self-reported

• DeepSWE

  #10 / 13

  DeepSWE is a software engineering agent benchmark evaluated with the mini-swe-agent harness, where each task is solved in an isolated container with no internet access. It measures an agent's ability to autonomously resolve real-world coding issues end to end.

  46.2%

  Self-reported

• HMMT Feb 26

  #6 / 13

  HMMT February 2026 is a math competition benchmark based on problems from the Harvard-MIT Mathematics Tournament, testing advanced mathematical problem-solving and reasoning.

  92.5%

  Self-reported

• SWE-Marathon

  #10 / 10

  SWE-Marathon is an ultra-long-horizon software engineering benchmark covering tasks such as building compilers, optimizing kernels, and developing production-grade services. It measures whether agents can sustain quality across extremely long engineering trajectories.

  13.0%

  Self-reported

• PostTrainBench

  #7 / 9

  PostTrainBench evaluates a model's ability to autonomously post-train base models. Given pretrain-only base models, the agent must complete the full pipeline of data synthesis, training, evaluation, and iteration within a time budget, scored across downstream benchmarks such as AIME2025, BFCL, GPQA Main, GSM8K, and HumanEval.

  34.3%

  Self-reported

• CritPT

  #2 / 6

  CritPT is a challenging reasoning benchmark reported by Qwen for evaluating frontier mathematical and critical problem-solving capability.

  16.7%

  Self-reported