Skip to main content

GLM-5.2 vs Mistral Large 4

SpecGLM-5.2Mistral Large 4
Respan Index43.4 (#61)44.1 (#59)
Blended price per 1M$0.593$1.05
Context window1M1M
Input / output per 1M$0.19 / $1.8$0.7 / $2.1
Max output131K-
Inputtexttext, vision
Released2026-06-242026-10-06

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryGLM-5.2Mistral Large 4
Coding27.9 GLM-5.237.3 Mistral Large 4
Agents55.8 GLM-5.256.7 Mistral Large 4
Reasoning42.0 GLM-5.242.1 Mistral Large 4
Knowledge & language34.6 GLM-5.218.2 Mistral Large 4
Human preference56.7 GLM-5.255.3 Mistral Large 4

How the Respan Index works

Benchmarks both report

GLM-5.2 leads on 10 of the 19 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkGLM-5.2Mistral Large 4
LMArena Agent (LMArena)-0.0041-0.0715
LMArena Elo14751422
LMArena WebDev (LMArena)16081568
MirrorCode (Epoch AI)36.7%40%
Chess Puzzles (Epoch AI)22%33%
APEX-Agents (Mercor)45.2%51.6%
FrontierMath Tier 4 v2 (Epoch AI)29.3%26.8%
FrontierMath Tiers 1-3 v2 (Epoch AI)59.2%58.6%
GPQA Diamond91.9%88.1%
LiveBench Agentic Coding (LiveBench)51.8%57.2%
LiveBench Coding (LiveBench)79.7%77.2%
LiveBench Data Analysis (LiveBench)73.7%76.5%
LiveBench Instruction Following (LiveBench)62.3%64.8%
LiveBench Language (LiveBench)76.2%49.6%
LiveBench Mathematics (LiveBench)89.8%93.6%
LiveBench Reasoning (LiveBench)78.6%83.9%
Mystery Game Puzzles (Epoch AI)19%17%
OTIS Mock AIME 2024-2025 (Epoch AI)86.4%98.3%
SimpleQA Verified34.2%20%

FAQ

Is GLM-5.2 better than Mistral Large 4?

Mistral Large 4 scores 44.1 on the Respan Index (#59), ahead of GLM-5.2 at 43.4 (#61). GLM-5.2 leads on 10 of the 19 benchmarks both report.

Which is cheaper, GLM-5.2 or Mistral Large 4?

GLM-5.2 lists at $0.19 input and $1.8 output per 1M tokens; Mistral Large 4 at $0.7 and $2.1. GLM-5.2 is 1.8x cheaper per token at list price.

Which has the larger context window?

Both accept 1M tokens of context.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.