Model comparison
GLM-5 vs GPT-4o mini
GLM-5 is the stronger model overall, scoring 46.1 to 25.5 on the Noometry Index. GPT-4o mini costs 5.9× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. GLM-5 scores higher in 9 categories and GPT-4o mini in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5 leads 46.4 to 10.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 80% for GLM-5 and 6.9% for GPT-4o mini.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GLM-5 accepts more context: 205K tokens versus 128K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | GPT-4o mini | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 25.5 |
| Released | 2026-02-11 | 2024-07-18 |
| Weights | Open | Proprietary |
| Context window | 205K | 128K |
| Max output | 131K | 16K |
| Input $ / M tokens | $1 | $0.15 |
| Output $ / M tokens | $3.20 | $0.60 |
| Results tracked | 45 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#52), GPT-4o mini: 22.0 (#335)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| WeirdML | 48.2% | 11.8% |
| LMArena Coding | 1461 | 1290 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| Aider Polyglot | — | 3.6% |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| BigCodeBench Instruct | — | 46.1% |
| LiveBench Coding | — | 43.1% |
| BigCodeBench Complete | — | 57.4% |
| ALE-Bench | 765.62 | — |
| HumanEval+ | — | 83.5% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use GLM-5 leads
GLM-5: 31.1 (#71), GPT-4o mini: 27.5 (#101)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| BALROG | — | 17.4% |
| Vending-Bench 2 | 4,432 | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), GPT-4o mini: 8.7 (#347)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| ARC-AGI-2 | 4.9% | 0% |
| SimpleBench | 53.2% | 10.7% |
| Kagi LLM Benchmark | 75% | 28.8% |
| Chess Puzzles | 10% | 0% |
| LMArena Hard Prompts | 1452 | 1267 |
| Epoch Capabilities Index | 145.83 | 126.56 |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| LiveBench Reasoning | — | 32.8% |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 54.4% |
| LiveBench Data Analysis | — | 50% |
| LMCA | — | 10.4% |
| ForecastBench | 61 | — |
| LiveBench | — | 41.3% |
| PIQA | — | 88.7% |
Math GLM-5 leads
GLM-5: 46.4 (#71), GPT-4o mini: 10.4 (#314)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 6.9% |
| LMArena Math | 1440 | 1267 |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 28% |
| LiveBench Math | — | 36.3% |
| MATH Level 5 | — | 52.6% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
| GSM8K | — | 91.3% |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), GPT-4o mini: 17.7 (#284)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 87.8% | 37.7% |
| LMArena Expert | 1454 | 1235 |
| SimpleQA Verified | — | 8.3% |
| MMLU-Pro | — | 60.3% |
| Confabulations | — | 37.2% |
| Vectara Hallucination Rate | 10.1% | — |
| GPQA (HELM) | — | 36.8% |
| BoolQ | — | 88.7% |
| MMLU | — | 81.8% |
Multimodal Not comparable
GLM-5: —, GPT-4o mini: 25.9 (#122)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| LMArena Vision | — | 1066 |
| Video-MME | — | 64.8% |
| GeoBench | — | 64% |
| VPCT | — | 34% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), GPT-4o mini: 42.0 (#199)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1430 | 1266 |
| LMArena Chinese | 1511 | 1265 |
| LMArena French | 1455 | 1297 |
| LMArena German | 1445 | 1272 |
| LMArena Japanese | 1416 | 1216 |
| LMArena Korean | 1423 | 1195 |
| LMArena Russian | 1436 | 1275 |
| LMArena Spanish | 1454 | 1276 |
Instruction Following GLM-5 leads
GLM-5: 75.2 (#67), GPT-4o mini: 61.9 (#239)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| LMArena Instruction Following | 1428 | 1258 |
| LiveBench Instruction Following | — | 56.8% |
| IFEval | — | 78.2% |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), GPT-4o mini: 39.1 (#186)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1446 | 1289 |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), GPT-4o mini: 39.5 (#248)
| Benchmark | GLM-5 | GPT-4o mini |
|---|---|---|
| LMArena Text | 1446 | 1286 |
| LMArena Creative Writing | 1439 | 1268 |
| EQ-Bench Creative Writing | 1601 | 873 |
| LMArena Multi-Turn | 1456 | 1285 |
| Short-Story Creative Writing | — | 67.2% |
| WildBench | — | 79.1% |
| LiveBench Language | — | 28.6% |
Frequently asked questions
Is GLM-5 better than GPT-4o mini?
GLM-5 is the stronger model overall, scoring 46.1 to 25.5 on the Noometry Index. GPT-4o mini costs 5.9× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Which is cheaper, GLM-5 or GPT-4o mini?
GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-5 or GPT-4o mini better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 22.0 in the Noometry coding category.
Which has the bigger context window?
GLM-5 does, with 205K tokens against 128K.
How many benchmarks do GLM-5 and GPT-4o mini share?
26 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and GPT-4o mini has 60.