Compare / head-to-head
GLM-5.2vs
GPT-5.1-Codex Mini
GLM-5.2 leads 1 of 1 shared benchmark. GPT-5.1-Codex Mini is 2.6x cheaper per token. GLM-5.2 has the larger context window (1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 0
GLM-5.2 leads
Cheaper per token
GPT-5.1-Codex Mini
2.6x cheaper, input + output
Larger context
GLM-5.2
1M tokens
Providers
2 providers
Z.ai · OpenAI
GLM-5.2
Z.ai · released 2026-06-24
text
GPT-5.1-Codex Mini
OpenAI · released 2025-11-13
textvision
Quality
Benchmark matrix
1 shared · 45 only GLM-5.2 · 0 only GPT-5.1-Codex Mini| Benchmark | GLM-5.2 | GPT-5.1-Codex Mini | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 1 | ||||
| LMArena WebDev (LMArena) | 1605 ↗ | 1243.6 ↗ | +361.4 | GLM-5.2 |
| Only GLM-5.2 reports · 45 | ||||
| GPQA Diamond | 91.9% ↗epoch run | not reported | — | — |
| AIME 2026 | 90% ↗matharena ⚠ | not reported | — | — |
| LMArena Elo | 1475.8 ↗ | not reported | — | — |
| Agents' Last Exam (ALE-CLI) | 23.8% ↗ | not reported | — | — |
| APEX-Agents (Mercor) | 45.2% ↗ | not reported | — | — |
| AutomationBench v1.0.6 | 26.2% ↗ | not reported | — | — |
| Chess Puzzles (Epoch AI run) | 21% ↗epoch run | not reported | — | — |
| CyberGym | 77.2% ↗ | not reported | — | — |
| DeepSWE v1.1 | 46.2% ↗ | not reported | — | — |
| DeepSWE v1.1 (Datacurve) | 43.8% ↗ | not reported | — | — |
| EBR-bench (Epoch AI run) | 9.5% ↗epoch run | not reported | — | — |
| ExploitBench | 24.4% ↗ | not reported | — | — |
| ExploitGym 2h | 29 ↗ | not reported | — | — |
| ExploitGym 6h | 39 ↗ | not reported | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | 29.3% ↗epoch run | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 59.2% ↗epoch run | not reported | — | — |
| FrontierSWE | 67.5% ↗ | not reported | — | — |
| GDPval-AA v2 | 1508 ↗ | not reported | — | — |
| HLE (with tools) | 54.7% ↗ | not reported | — | — |
| LiveBench Agentic Coding (LiveBench) | 51.8% ↗ | not reported | — | — |
| LiveBench Coding (LiveBench) | 79.7% ↗ | not reported | — | — |
| LiveBench Data Analysis (LiveBench) | 73.7% ↗ | not reported | — | — |
| LiveBench Instruction Following (LiveBench) | 62.3% ↗ | not reported | — | — |
| LiveBench Language (LiveBench) | 76.2% ↗ | not reported | — | — |
| LiveBench Mathematics (LiveBench) | 89.8% ↗ | not reported | — | — |
| LiveBench Reasoning (LiveBench) | 78.6% ↗ | not reported | — | — |
| LMArena Agent (LMArena) | 0.0348 ↗ | not reported | — | — |
| Mystery Game Puzzles (Epoch AI run) | 19% ↗epoch run | not reported | — | — |
| NL2Repo | 48.9% ↗ | not reported | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 86.4% ↗epoch run | not reported | — | — |
| PostTrainBench | 31.7% ↗ | not reported | — | — |
| ProgramBench (Almost Solved) | 9.5% ↗ | not reported | — | — |
| SimpleBench (SimpleBench) | 58.8% ↗ | not reported | — | — |
| SimpleQA Verified | 34.2% ↗epoch run | not reported | — | — |
| SWE-Bench Pro | 62.1% ↗ | not reported | — | — |
| SWE-bench Verified (Epoch AI run) | 78.7% ↗epoch run | not reported | — | — |
| SWE-Marathon v1.1 | 19.4% ↗ | not reported | — | — |
| SWE-rebench 2026-05-15 to 2026-07-01 (Nebius) | 62.9% ↗ | not reported | — | — |
| tau2-bench Banking Knowledge (Sierra) | 37.1% ↗ | not reported | — | — |
| Terminal-Bench 2.1 | 81% ↗ | not reported | — | — |
| Terminal-Bench 3.0 | 4.6% ↗ | not reported | — | — |
| Toolathlon Verified | 59.9% ↗ | not reported | — | — |
| Toolathlon-Verified (HKUST) | 59.9% ↗ | not reported | — | — |
| Vending-Bench 2 (Andon Labs) | 8313.78 ↗ | not reported | — | — |
| WeirdML (Håvard Tveit Ihle) | 70.1% ↗ | not reported | — | — |
| Only GPT-5.1-Codex Mini reports · 0 | ||||
| GPT-5.1-Codex Mini reports nothing GLM-5.2 does not. | ||||
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GLM-5.2 minus GPT-5.1-Codex Mini in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | GLM-5.2 | GPT-5.1-Codex Mini | Edge |
|---|---|---|---|
| Context window | 1M tokens | 400K tokens | GLM-5.2 |
| Max output | 131K tokens | 128K tokens | GLM-5.2 |
| Input price / 1M | $1.4 ↗ | $0.25 ↗ | GPT-5.1-Codex Mini |
| Output price / 1M | $4.4 ↗ | $2 ↗ | GPT-5.1-Codex Mini |
| Cached input / 1M | $0.26 ↗ | $0.025 ↗ | GPT-5.1-Codex Mini |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $5.8 | $2.25 | GPT-5.1-Codex Mini |
| Modalities | text | text · vision | GPT-5.1-Codex Mini |
| Released | 2026-06-24 | 2025-11-13 | — |
| Cited benchmark scores | 53 | 1 | — |
Reliability
Provider status
Live from /statusMore matchups
GLM-5.2 vs …
Models sharing the most benchmarksMore matchups
GPT-5.1-Codex Mini vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run GLM-5.2 and GPT-5.1-Codex Mini on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.