Z.aiEco TierMoE SparseZero Data Retention

GLM-5.2

Name: ARMES
Availability: InStock
Author: ARMES

Run GLM-5.2

Included on ECO plan

Context Window

1M (solid/stable)

Input Token Price

$1.40

per 1,000,000 tokens

Output Token Price

$4.40

per 1,000,000 tokens

Speed Rating

Moderate

Verified Frontier Benchmarks

Empirical Evaluation Results

Evaluated across standard and verified harnesses

SWE-Bench Pro

coding

62.1 (#1 open model; beats GPT-5.5 58.6, GLM-5.1 58.4)

Terminal-Bench 2.1

agentic

81.0 (vs GLM-5.1's 62.0; near Opus 4.8's 85.0, beats Gemini 3.1 Pro)

FrontierSWE

coding

74.4 (beats GPT-5.5 72.6, ~tie Opus 4.8 75.1)

MCP-Atlas

reasoning

77.0 (beats GPT-5.5 75.3)

HLE w/ Tools

reasoning

54.7 (beats GPT-5.5 52.2)

AIME 2026

math

99.2 (beats Opus 4.8 & GPT-5.5)

PostTrainBench

reasoning

34.3 (vs GPT-5.5 25.0). Code Arena Frontend: #2 overall; Agent Arena: #1 open model

Architectural Profile & Capabilities

Long-horizon agentic engineering, repository-scale refactors, multi-hour autonomous coding, cross-file/long-chain tasks, frontend coding (best-in-class among open models), tool use, and competition math. The strongest open-weight coding/agent model available — trails only Claude Opus 4.8, at ~1/6th the cost. MoE 744B/40B-active, MIT-licensed open weights. Dual reasoning-effort modes (`high` / `max`). Released Jun 13, 2026. *Superseded by GLM-5.3.*

Architecture Type: Mixture of Experts (MoE)

Sparse Mixture-of-Experts routing: activates only a subset of parameter experts per token, delivering frontier-level intelligence with exceptional efficiency.

Recommended Workloads & Primary Use Cases

Everyday reasoning

Drafting

Fast problem solving

ARMES Zero Data Retention

Calls to GLM-5.2 are routed through strict Zero Data Retention inference channels. Prompts and outputs are never stored, indexed, or monitored by Z.ai.

View Privacy Architecture

More Models from Z.ai

Compare other engines available in this laboratory.

View Roster (6)

GLM-5.3 FlashFree Tier

Zhipu AI's 320B/18B-active multimodal MoE model featuring hybrid linear attention, 1.31M context, and top-tier agentic coding efficiency.

1.31M ctx$0.15/M

GLM-5.3Eco Tier

Z.ai's most capable model (released Aug 14, 2026).

1M ctx$1.40/M

GLM-5.1Eco Tier

GLM-5.1 is a high-performance foundation model engineered by Z.ai, accessible with Zero Data Retention on ARMES.

128K ctx$1.40/M

Open App

Back to Z.ai models roster

Z.aiEco TierMoE SparseZero Data Retention

GLM-5.2

Run GLM-5.2

Included on ECO plan

Context Window

1M (solid/stable)

Input Token Price

$1.40

per 1,000,000 tokens

Output Token Price

$4.40

per 1,000,000 tokens

Speed Rating

Moderate

Verified Frontier Benchmarks

Empirical Evaluation Results

Evaluated across standard and verified harnesses

SWE-Bench Pro

coding

62.1 (#1 open model; beats GPT-5.5 58.6, GLM-5.1 58.4)

Terminal-Bench 2.1

agentic

81.0 (vs GLM-5.1's 62.0; near Opus 4.8's 85.0, beats Gemini 3.1 Pro)

FrontierSWE

coding

74.4 (beats GPT-5.5 72.6, ~tie Opus 4.8 75.1)

MCP-Atlas

reasoning

77.0 (beats GPT-5.5 75.3)

HLE w/ Tools

reasoning

54.7 (beats GPT-5.5 52.2)

AIME 2026

math

99.2 (beats Opus 4.8 & GPT-5.5)

PostTrainBench

reasoning

34.3 (vs GPT-5.5 25.0). Code Arena Frontend: #2 overall; Agent Arena: #1 open model

Architectural Profile & Capabilities

Architecture Type: Mixture of Experts (MoE)

Sparse Mixture-of-Experts routing: activates only a subset of parameter experts per token, delivering frontier-level intelligence with exceptional efficiency.

Recommended Workloads & Primary Use Cases

Everyday reasoning

Drafting

Fast problem solving

ARMES Zero Data Retention

Calls to GLM-5.2 are routed through strict Zero Data Retention inference channels. Prompts and outputs are never stored, indexed, or monitored by Z.ai.

View Privacy Architecture

More Models from Z.ai

Compare other engines available in this laboratory.

View Roster (6)

GLM-5.3 FlashFree Tier

Zhipu AI's 320B/18B-active multimodal MoE model featuring hybrid linear attention, 1.31M context, and top-tier agentic coding efficiency.

1.31M ctx$0.15/M

GLM-5.3Eco Tier

Z.ai's most capable model (released Aug 14, 2026).

1M ctx$1.40/M

GLM-5.1Eco Tier

GLM-5.1 is a high-performance foundation model engineered by Z.ai, accessible with Zero Data Retention on ARMES.

128K ctx$1.40/M