Z.ai · open weights on Arena · first on Arena 16 June 2026

glm-5.2 (max)

130 days for open weights to match claude-opus-4-6

This open-weights model matched the score of claude-opus-4-6 in WebDev 130 days after claude-opus-4-6 took first place. It also matched claude-opus-4-6-thinking in WebDev, claude-opus-4-7 in WebDev and claude-opus-4-7-thinking in WebDev.

Embed

Who held first place

<iframe src="https://jesseahawkins.com/frontier/embed/" title="Frontier: who held first place on each Arena leaderboard, and for how long" width="100%" height="560" style="border:0;max-width:100%" loading="lazy"></iframe>

Credits LMArena.

The open-weights catch-up

<iframe src="https://jesseahawkins.com/frontier/embed/catch-up/" title="Frontier: the days until open weights matched each closed leader" width="100%" height="900" style="border:0;max-width:100%" loading="lazy"></iframe>

Credits LMArena.

The saturation clock

<iframe src="https://jesseahawkins.com/frontier/embed/clock/" title="Frontier: the days from each benchmark’s release to a model at 90%" width="100%" height="640" style="border:0;max-width:100%" loading="lazy"></iframe>

Credits Epoch AI.

Each frame links back to its page. Preview it.

Cite

This page

Jesse Hawkins, “glm-5.2 (max) · Frontier”, jesseahawkins.com, data retrieved 30 Oct 2026. https://jesseahawkins.com/frontier/glm-5-2-max/. Data: LMArena, Arena leaderboard dataset, revision 4491b06, CC BY 4.0; Epoch AI, Capabilities & benchmarking, CC BY 4.0.

BibTeX

@misc{hawkins2026frontierglm52max,
  author = {Hawkins, Jesse},
  title = {{glm-5.2 (max) · Frontier}},
  year = {2026},
  howpublished = {\url{https://jesseahawkins.com/frontier/glm-5-2-max/}},
  note = {Data: LMArena, Arena leaderboard dataset, revision 4491b06, CC BY 4.0; Epoch AI, Capabilities \& benchmarking, CC BY 4.0, retrieved 2026-10-30}
}

Arena · LMArena

LMArena, "Arena leaderboard dataset", Hugging Face, revision 4491b06. https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset. CC BY 4.0.

Benchmarks · Epoch AI

Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks' [online resource]. CC BY 4.0.

Download

Post on X

Text: best place 16

last on the board 21 July 2026

1st10th

On the board: 16 June 2026 to 21 July 2026
Score: 1464 (1458–1470) on 21 July 2026 · 18,017 votes

WebDev: best place 2

last on the board 24 July 2026

1st10th

On the board: 16 June 2026 to 24 July 2026
Score: 1588 (1578–1597) on 24 July 2026 · 5,537 votes
Matched: claude-opus-4-6 on 16 June 2026, 130 days after it took first place
claude-opus-4-6-thinking on 16 June 2026, 127 days after it took first place
claude-opus-4-7 on 16 June 2026, 60 days after it took first place
claude-opus-4-7-thinking on 16 June 2026, 57 days after it took first place

Epoch AI: GLM-5.2

released 16 June 2026 · Open weights (unrestricted)

Chess Puzzles 22.0%
EBR-bench 9.5%
FrontierMath-Tier-4-v2-Private 29.3%
FrontierMath-Tiers-1-3-v2-Private 59.2%
GPQA diamond 91.9%
Mystery Game Puzzles 19.0%
OTIS Mock AIME 2024-2025 86.4%
SWE-Bench verified 78.7%
SimpleQA Verified 34.2%

Sources

Arena scores come from people voting between two anonymous answers. First place is the highest score; the top two are often within each other’s margin of error. LMArena, Arena leaderboard dataset, revision 4491b06, retrieved 30 October 2026; Epoch AI, Capabilities & benchmarking, retrieved 30 October 2026. Both CC BY 4.0.