GLM-5.2 vs Qwen3.7-Max
| Spec | GLM-5.2 | Qwen3.7-Max |
|---|---|---|
| Respan Index | 42.1 (#54) | 43.1 (#52) |
| Blended price per 1M | $0.354 | $1.88 |
| Context window | 1M | 1M |
| Input / output per 1M | $0.006 / $1.4 | $1.25 / $3.75 |
| Max output | 131K | - |
| Input | text | text, vision |
| Released | 2026-06-24 | 2026-05-20 |
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.
| Category | GLM-5.2 | Qwen3.7-Max |
|---|---|---|
| Coding | 22.8 GLM-5.2 | 16.6 Qwen3.7-Max |
| Agents | 56.4 GLM-5.2 | 53.0 Qwen3.7-Max |
| Reasoning | 41.9 GLM-5.2 | 48.5 Qwen3.7-Max |
| Knowledge & language | 34.5 GLM-5.2 | 55.0 Qwen3.7-Max |
| Human preference | 56.1 GLM-5.2 | 53.7 Qwen3.7-Max |
Benchmarks both report
Qwen3.7-Max leads on 10 of the 19 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | GLM-5.2 | Qwen3.7-Max |
|---|---|---|
| Chess Puzzles (Epoch AI) | 22% | 26% |
| LMArena Agent (LMArena) | 0.0126 | -0.0659 |
| LMArena WebDev (LMArena) | 1606 | 1515 |
| EBR-bench (Epoch AI) | 9.5% | 9.5% |
| FrontierMath Tier 4 v2 (Epoch AI) | 29.3% | 34.1% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 59.2% | 64.6% |
| GPQA Diamond | 91.9% | 90.9% |
| LiveBench Agentic Coding (LiveBench) | 51.8% | 43.6% |
| LiveBench Coding (LiveBench) | 79.7% | 74.2% |
| LiveBench Data Analysis (LiveBench) | 73.7% | 71.8% |
| LiveBench Instruction Following (LiveBench) | 62.3% | 74% |
| LiveBench Language (LiveBench) | 76.2% | 79.7% |
| LiveBench Mathematics (LiveBench) | 89.8% | 85.2% |
| LiveBench Reasoning (LiveBench) | 78.6% | 83.3% |
| Mystery Game Puzzles (Epoch AI) | 19% | 32% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 86.4% | 95.6% |
| SimpleQA Verified | 34.2% | 55.8% |
| SWE-bench Verified (Epoch AI) | 78.7% | 77.3% |
| SimpleBench (SimpleBench) | 58.8% | 70.4% |
FAQ
Is GLM-5.2 better than Qwen3.7-Max?
Qwen3.7-Max scores 43.1 on the Respan Index (#52), ahead of GLM-5.2 at 42.1 (#54). Qwen3.7-Max leads on 10 of the 19 benchmarks both report.
Which is cheaper, GLM-5.2 or Qwen3.7-Max?
GLM-5.2 lists at $0.006 input and $1.4 output per 1M tokens; Qwen3.7-Max at $1.25 and $3.75. GLM-5.2 is 5.3x cheaper per token at list price.
Which has the larger context window?
Both accept 1M tokens of context.