Model comparison
GLM-5 vs gpt-oss-120b
GLM-5 is the stronger model overall, scoring 46.1 to 36.3 on the Noometry Index. gpt-oss-120b costs 22× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GLM-5 scores higher in 8 categories and gpt-oss-120b in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5 leads 66.0 to 46.5.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 72.8% for GLM-5 and 26% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GLM-5 accepts more context: 205K tokens versus 131K.
Side by side
| GLM-5 | gpt-oss-120b | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 36.3 |
| Released | 2026-02-11 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 205K | 131K |
| Max output | 131K | 41K |
| Input $ / M tokens | $1 | $0.037 |
| Output $ / M tokens | $3.20 | $0.17 |
| Results tracked | 45 | 48 |
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-oss-120b: 33.5 (#256)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| SWE-bench Verified (bash only) | 72.8% | 26% |
| WeirdML | 48.2% | 48.2% |
| LMArena Coding | 1461 | 1380 |
| ALE-Bench | 765.62 | 575.62 |
| SWE-bench Verified | 72.1% | — |
| Aider Polyglot | — | 41.8% |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 36% |
| AlgoTune | — | 1.41 |
Agentic & Tool Use GLM-5 leads
GLM-5: 31.1 (#71), gpt-oss-120b: 12.2 (#153)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | 52.4% | 18.7% |
| Vending-Bench 2 | 4,432 | -21.53 |
| APEX-Agents | — | 4.4% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| METR Time Horizons | — | 56.6% |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), gpt-oss-120b: 20.0 (#245)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| SimpleBench | 53.2% | 22.1% |
| Kagi LLM Benchmark | 75% | 58.6% |
| Chess Puzzles | 10% | 20% |
| LMArena Hard Prompts | 1452 | 1364 |
| Epoch Capabilities Index | 145.83 | 139.93 |
| ARC-AGI-2 | 4.9% | — |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| CritPt | — | 1.1% |
| Mystery Game Puzzles | — | 2% |
| DTBench | — | 76.3% |
| LMCA | — | 22.1% |
| Surface Evolver Bench | — | 25% |
| ForecastBench | 61 | — |
Math gpt-oss-120b leads
GLM-5: 46.4 (#71), gpt-oss-120b: 52.5 (#50)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 88.9% |
| LMArena Math | 1440 | 1389 |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 68.8% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), gpt-oss-120b: 42.4 (#96)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 87.8% | 75.8% |
| Vectara Hallucination Rate | 10.1% | 14.2% |
| LMArena Expert | 1454 | 1356 |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| GPQA (HELM) | — | 68.4% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), gpt-oss-120b: 48.0 (#147)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1430 | 1351 |
| LMArena Chinese | 1511 | 1385 |
| LMArena French | 1455 | 1369 |
| LMArena German | 1445 | 1353 |
| LMArena Japanese | 1416 | 1331 |
| LMArena Korean | 1423 | 1282 |
| LMArena Russian | 1436 | 1343 |
| LMArena Spanish | 1454 | 1389 |
Instruction Following GLM-5 leads
GLM-5: 75.2 (#67), gpt-oss-120b: 69.3 (#173)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1428 | 1318 |
| IFEval | — | 83.6% |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), gpt-oss-120b: 31.4 (#278)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1446 | 1319 |
| Fiction.LiveBench | — | 44.4% |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), gpt-oss-120b: 46.5 (#217)
| Benchmark | GLM-5 | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1446 | 1365 |
| LMArena Creative Writing | 1439 | 1275 |
| EQ-Bench Creative Writing | 1601 | 961 |
| LMArena Multi-Turn | 1456 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| WildBench | — | 84.5% |
Frequently asked questions
Is GLM-5 better than gpt-oss-120b?
GLM-5 is the stronger model overall, scoring 46.1 to 36.3 on the Noometry Index. gpt-oss-120b costs 22× 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-oss-120b?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-5 or gpt-oss-120b better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
GLM-5 does, with 205K tokens against 131K.
How many benchmarks do GLM-5 and gpt-oss-120b share?
30 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and gpt-oss-120b has 48.