Gemini 3.1 Pro Preview vs GLM-5.2
| Spec | Gemini 3.1 Pro Preview | GLM-5.2 |
|---|---|---|
| Respan Index | 43.6 (#60) | 43.4 (#61) |
| Blended price per 1M | $4.5 | $0.593 |
| Context window | 1.0M | 1M |
| Input / output per 1M | $2 / $12 | $0.19 / $1.8 |
| Max output | 66K | 131K |
| Input | text, vision, audio | text |
| Released | 2026-02-19 | 2026-06-24 |
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 | Gemini 3.1 Pro Preview | GLM-5.2 |
|---|---|---|
| Coding | 14.7 Gemini 3.1 Pro Preview | 27.9 GLM-5.2 |
| Agents | 53.4 Gemini 3.1 Pro Preview | 55.8 GLM-5.2 |
| Reasoning | 45.8 Gemini 3.1 Pro Preview | 42.0 GLM-5.2 |
| Knowledge & language | 72.5 Gemini 3.1 Pro Preview | 34.6 GLM-5.2 |
| Human preference | 52.7 Gemini 3.1 Pro Preview | 56.7 GLM-5.2 |
Benchmarks both report
Gemini 3.1 Pro Preview leads on 17 of the 29 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Gemini 3.1 Pro Preview | GLM-5.2 |
|---|---|---|
| LMArena Agent (LMArena) | -0.0569 | -0.0041 |
| LMArena Elo | 1487 | 1475 |
| LMArena WebDev (LMArena) | 1446 | 1608 |
| APEX-Agents (Mercor) | 35.1% | 45.2% |
| Chess Puzzles (Epoch AI) | 51% | 22% |
| DeepSWE v1.1 (Datacurve) | 11.7% | 43.8% |
| EBR-bench (Epoch AI) | 14.3% | 9.5% |
| FrontierMath Tier 4 v2 (Epoch AI) | 26.8% | 29.3% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 59.6% | 59.2% |
| GPQA Diamond | 94.3% | 91.9% |
| LiveBench Agentic Coding (LiveBench) | 44.1% | 51.8% |
| LiveBench Coding (LiveBench) | 76.5% | 79.7% |
| LiveBench Data Analysis (LiveBench) | 78.5% | 73.7% |
| LiveBench Instruction Following (LiveBench) | 79.1% | 62.3% |
| LiveBench Language (LiveBench) | 85.4% | 76.2% |
| LiveBench Mathematics (LiveBench) | 91% | 89.8% |
| LiveBench Reasoning (LiveBench) | 84% | 78.6% |
| MirrorCode (Epoch AI) | 8.9% | 36.7% |
| Mystery Game Puzzles (Epoch AI) | 34% | 19% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 95.6% | 86.4% |
| SimpleQA Verified | 73.5% | 34.2% |
| SWE-bench Verified (Epoch AI) | 75.6% | 78.7% |
| Vending-Bench 2 (Andon Labs) | 3774.25 | 8313.78 |
| WeirdML (Håvard Tveit Ihle) | 72.1% | 70.1% |
| AIME 2026 | 98.3% | 90% |
| SWE-Bench Pro | 54.2% | 62.1% |
| Toolathlon-Verified (HKUST) | 61.1% | 59.9% |
| tau2-bench Banking Knowledge (Sierra) | 26% | 37.1% |
| SimpleBench (SimpleBench) | 79.6% | 58.8% |
FAQ
Is Gemini 3.1 Pro Preview better than GLM-5.2?
Gemini 3.1 Pro Preview scores 43.6 on the Respan Index (#60), ahead of GLM-5.2 at 43.4 (#61). Gemini 3.1 Pro Preview leads on 17 of the 29 benchmarks both report.
Which is cheaper, Gemini 3.1 Pro Preview or GLM-5.2?
Gemini 3.1 Pro Preview lists at $2 input and $12 output per 1M tokens; GLM-5.2 at $0.19 and $1.8. GLM-5.2 is 7.6x cheaper per token at list price.
Which has the larger context window?
Gemini 3.1 Pro Preview accepts 1.0M tokens and GLM-5.2 1M tokens.
Compare other models
- vsGemini 3.1 Pro Preview vs Gemini 3 Pro Preview
- vsGLM-5.2 vs GLM-5.1
- vsGLM-5.3 vs GLM-5.2
- vsClaude Sonnet 5.5 vs Gemini 3.1 Pro Preview
- vsGPT-5.6 Luna vs Gemini 3.1 Pro Preview
- vsGemini 3.1 Pro Preview vs Mistral Large 4
- vsGemini 3.1 Pro Preview vs Gemini 3.5 Flash
- vsGemini 3.1 Pro Preview vs Qwen3.7-Max
- vsGLM-5.2 vs Qwen3.7-Max
- vsGLM-5.2 vs Mistral Large 4
- vsGLM-5.2 vs Muse Spark Open