AI Models

Reasoning benchmarks

HLE, CritPt, AIME 2026, HMMT, GPQA-Diamond, IMOAnswerBench.

← Back to GLM-5.2 Benchmarks

About this data

GLM-5.2 is highly competitive on reasoning — it tops the field on AIME 2026 (99.2) and IMOAnswerBench, and sits close to the frontier on GPQA-Diamond and HMMT. HLE remains the hardest gap to the closed leaders. Some competitor HLE/CritPt cells are full-set scores, as noted in the source footnotes.

Reasoning benchmarks

Score (higher = better). GLM-5.2 leads on AIME 2026; competitive across the board.

View data & sources →

Data table

Reasoning benchmarks — reasoning data table (GLM-5.2 Benchmarks)
glm51 glm52 gpt55 gemini opus48 series benchmark source_ref value_basis
31 40.5 41.4 45 49.8 reasoning HLE zai-glm52 Full Benchmark Table — HLE (Opus/GPT marked full-set)
52.3 54.7 52.2 51.4 57.9 reasoning HLE w/ Tools zai-glm52 Full Benchmark Table — HLE with tools
4.6 20.9 27.1 17.7 20.9 reasoning CritPt zai-glm52 Full Benchmark Table — CritPt
95.3 99.2 98.3 98.2 95.7 reasoning AIME 2026 zai-glm52 Full Benchmark Table — AIME 2026
82.6 92.5 96.7 87.3 96.7 reasoning HMMT Feb 2026 zai-glm52 Full Benchmark Table — HMMT Feb 2026

2 more rows + CSV download

The full 7-row dataset, one-click CSV export, and the AI-ready context file are free with an account. Prefer to verify it yourself? The full methodology and sources are published below.

Methodology & sources

Last updated: Jul 17, 2026

Methodology

Source-backed values for all six charts come from the Z.ai GLM-5.2 launch post (2026-06-16): long-horizon coding (FrontierSWE, PostTrainBench, SWE-Marathon), standard coding (Terminal-Bench 2.1, SWE-bench Pro, NL2Repo, DeepSWE, ProgramBench), reasoning (HLE, CritPt, AIME, HMMT, GPQA-Diamond, IMOAnswerBench), agentic (MCP-Atlas, Tool-Decathlon), the GLM-5.2 vs GLM-5.1 generational leap, and the MTP acceptance-length ablation. Every numeric point carries a sources[].ref and a value_basis naming the table row. Scores are reproduced verbatim from the Full Benchmark Table; cells the vendor published as "-" (not reported) are omitted, never estimated. CAVEAT: these are the model developer’s self-reported figures under their stated harnesses and prompts (see the post’s footnotes); some competitor HLE/CritPt cells are full-set scores. They are comparison-as-published, not an independent eval. Architecture context (not charted): 1M-token context (up from 200K), IndexShare cuts per-token indexer FLOPs ~2.9× at 1M, MIT-licensed open weights. Re-verified 2026-06-22.

Sources

Comparisons are informative, not definitive. See each source for definitions and limits.

Continue researching

How relevant was this information?