Model comparison
GLM-5 vs gpt-oss-20b
GLM-5 is the stronger model overall, scoring 46.1 to 32.5 on the Noometry Index. gpt-oss-20b costs 43× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. GLM-5 scores higher in 9 categories and gpt-oss-20b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5 leads 66.0 to 35.5.
- The biggest single-benchmark swing is Terminal-Bench: 52.4% for GLM-5 and 3.4% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 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-20b | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 32.5 |
| Released | 2026-02-11 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 205K | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $1 | $0.018 |
| Output $ / M tokens | $3.20 | $0.09 |
| Results tracked | 45 | 34 |
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-20b: 37.6 (#192)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| WeirdML | 48.2% | 40.9% |
| LMArena Coding | 1461 | 1306 |
| ALE-Bench | 765.62 | 566.05 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 34.4% |
Agentic & Tool Use GLM-5 leads
GLM-5: 31.1 (#71), gpt-oss-20b: 9.3 (#154)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | 52.4% | 3.4% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| Vending-Bench 2 | 4,432 | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), gpt-oss-20b: 19.3 (#261)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| Kagi LLM Benchmark | 75% | 53.2% |
| Chess Puzzles | 10% | 4% |
| LMArena Hard Prompts | 1452 | 1274 |
| Epoch Capabilities Index | 145.83 | 137.82 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| CritPt | — | 1.4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| ForecastBench | 61 | — |
Math GLM-5 leads
GLM-5: 46.4 (#71), gpt-oss-20b: 39.4 (#103)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 65.3% |
| LMArena Math | 1440 | 1317 |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 56.5% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), gpt-oss-20b: 34.6 (#195)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 87.8% | 60.8% |
| LMArena Expert | 1454 | 1258 |
| MMLU-Pro | — | 74% |
| Vectara Hallucination Rate | 10.1% | — |
| GPQA (HELM) | — | 59.4% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), gpt-oss-20b: 42.2 (#197)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1430 | 1268 |
| LMArena Chinese | 1511 | 1314 |
| LMArena German | 1445 | 1255 |
| LMArena Japanese | 1416 | 1244 |
| LMArena Korean | 1423 | 1236 |
| LMArena Russian | 1436 | 1278 |
| LMArena Spanish | 1454 | 1267 |
| LMArena French | 1455 | — |
Instruction Following GLM-5 leads
GLM-5: 75.2 (#67), gpt-oss-20b: 61.8 (#240)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1428 | 1236 |
| IFEval | — | 73.2% |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), gpt-oss-20b: 37.9 (#209)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1446 | 1250 |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), gpt-oss-20b: 35.5 (#265)
| Benchmark | GLM-5 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1446 | 1287 |
| LMArena Creative Writing | 1439 | 1201 |
| EQ-Bench Creative Writing | 1601 | 666 |
| LMArena Multi-Turn | 1456 | 1268 |
| WildBench | — | 73.7% |
Frequently asked questions
Is GLM-5 better than gpt-oss-20b?
GLM-5 is the stronger model overall, scoring 46.1 to 32.5 on the Noometry Index. gpt-oss-20b costs 43× 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-20b?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-5 or gpt-oss-20b better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 37.6 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-20b share?
25 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and gpt-oss-20b has 34.