Model comparison
GLM-5 vs GPT-5 Mini
GLM-5 is the stronger model overall, scoring 46.1 to 41.8 on the Noometry Index. GPT-5 Mini costs 2.3× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GLM-5 scores higher in 6 categories and GPT-5 Mini in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5 leads 66.0 to 55.2.
- The biggest single-benchmark swing is SWE-bench Multilingual: 69.7% for GLM-5 and 39.7% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GPT-5 Mini accepts more context: 400K tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | GPT-5 Mini | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 41.8 |
| Released | 2026-02-11 | 2025-08-07 |
| Weights | Open | Proprietary |
| Context window | 205K | 400K |
| Max output | 131K | 128K |
| Input $ / M tokens | $1 | $0.25 |
| Output $ / M tokens | $3.20 | $2 |
| 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-5 Mini: 40.1 (#146)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| SWE-bench Verified | 72.1% | 64.7% |
| SWE-bench Verified (bash only) | 72.8% | 59.8% |
| SWE-bench Multilingual | 69.7% | 39.7% |
| WeirdML | 48.2% | 52.7% |
| LMArena Coding | 1461 | 1406 |
| ALE-Bench | 765.62 | 799.77 |
| LMArena WebDev | 1434 | — |
| SciCode | — | 39.2% |
| AlgoTune | — | 1.38 |
Agentic & Tool Use Too close to call
GLM-5: 31.1 (#71), GPT-5 Mini: 31.1 (#70)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | 52.4% | 34.8% |
| Vending-Bench 2 | 4,432 | -31.18 |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), GPT-5 Mini: 23.9 (#168)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 4.9% | 4.4% |
| Kagi LLM Benchmark | 75% | 70.3% |
| ARC-AGI-1 | 44.7% | 54.3% |
| Chess Puzzles | 10% | 30% |
| LMArena Hard Prompts | 1452 | 1380 |
| Epoch Capabilities Index | 145.83 | 145.52 |
| ForecastBench | 61 | 61 |
| SimpleBench | 53.2% | — |
| NYT Connections (extended) | 74.8% | — |
| CritPt | — | 0% |
| EnigmaEval | — | 8.2% |
| Mystery Game Puzzles | — | 10% |
| DTBench | — | 80.5% |
| LMCA | — | 34.2% |
Math Too close to call
GLM-5: 46.4 (#71), GPT-5 Mini: 46.7 (#69)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 86.7% |
| LMArena Math | 1440 | 1378 |
| FrontierMath (Feb 2025 set) | 16.4% | 27.2% |
| FrontierMath Tier 4 (v1) | 2.1% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| MathArena Final-Answer Competitions | 65.7% | — |
| ProofBench | — | 9% |
| Omni-MATH | — | 72.2% |
| MATH Level 5 | — | 97.8% |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), GPT-5 Mini: 45.6 (#86)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 87.8% | 75% |
| Vectara Hallucination Rate | 10.1% | 12.9% |
| LMArena Expert | 1454 | 1379 |
| Humanity's Last Exam | — | 19.4% |
| SimpleQA Verified | — | 21.6% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| GPQA (HELM) | — | 75.6% |
Multimodal Not comparable
GLM-5: —, GPT-5 Mini: 35.6 (#85)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), GPT-5 Mini: 48.9 (#137)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1430 | 1363 |
| LMArena Chinese | 1511 | 1385 |
| LMArena French | 1455 | 1386 |
| LMArena German | 1445 | 1366 |
| LMArena Japanese | 1416 | 1341 |
| LMArena Korean | 1423 | 1308 |
| LMArena Russian | 1436 | 1362 |
| LMArena Spanish | 1454 | 1355 |
Instruction Following GPT-5 Mini leads
GLM-5: 75.2 (#67), GPT-5 Mini: 76.2 (#46)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1428 | 1357 |
| IFEval | — | 92.7% |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), GPT-5 Mini: 41.9 (#132)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1446 | 1355 |
| Fiction.LiveBench | — | 69.4% |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), GPT-5 Mini: 55.2 (#148)
| Benchmark | GLM-5 | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1446 | 1373 |
| LMArena Creative Writing | 1439 | 1325 |
| EQ-Bench Creative Writing | 1601 | 1313 |
| LMArena Multi-Turn | 1456 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |
Frequently asked questions
Is GLM-5 better than GPT-5 Mini?
GLM-5 is the stronger model overall, scoring 46.1 to 41.8 on the Noometry Index. GPT-5 Mini costs 2.3× 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-5 Mini?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-5 or GPT-5 Mini better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 205K.
How many benchmarks do GLM-5 and GPT-5 Mini share?
36 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and GPT-5 Mini has 60.