Model comparison
GLM-4.6 vs gpt-oss-20b
GLM-4.6 is the stronger model overall, scoring 41.4 to 32.5 on the Noometry Index. gpt-oss-20b costs 28× less per token, which makes it the better buy when GLM-4.6's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Summary
- They share 22 benchmarks with published results for both. GLM-4.6 scores higher in 8 categories and gpt-oss-20b in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.6 leads 61.1 to 35.5.
- The biggest single-benchmark swing is Terminal-Bench: 24.5% for GLM-4.6 and 3.4% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.60 / $2.20 for GLM-4.6.
- GLM-4.6 accepts more context: 205K tokens versus 131K.
Side by side
| GLM-4.6 | gpt-oss-20b | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 41.4 | 32.5 |
| Released | 2025-09-30 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 205K | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $0.60 | $0.018 |
| Output $ / M tokens | $2.20 | $0.09 |
| Results tracked | 29 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.6 leads
GLM-4.6: 40.1 (#148), gpt-oss-20b: 37.6 (#192)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| SciCode | 38.4% | 34.4% |
| LMArena Coding | 1449 | 1306 |
| ALE-Bench | 340.82 | 566.05 |
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1340 | — |
| WeirdML | — | 40.9% |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.3 (#66), gpt-oss-20b: 9.3 (#154)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | 24.5% | 3.4% |
| Berkeley Function Calling Leaderboard | 72.4% | — |
Reasoning GLM-4.6 leads
GLM-4.6: 23.7 (#172), gpt-oss-20b: 19.3 (#261)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | 53.2% |
| CritPt | 1.1% | 1.4% |
| LMArena Hard Prompts | 1440 | 1274 |
| Chess Puzzles | — | 4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| Epoch Capabilities Index | — | 137.82 |
Math Too close to call
GLM-4.6: 39.1 (#111), gpt-oss-20b: 39.4 (#103)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Math | 1432 | 1317 |
| OTIS Mock AIME 2024-2025 | — | 65.3% |
| Omni-MATH | — | 56.5% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-4.6 leads
GLM-4.6: 40.2 (#124), gpt-oss-20b: 34.6 (#195)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Expert | 1431 | 1258 |
| GPQA Diamond | — | 60.8% |
| MMLU-Pro | — | 74% |
| Vectara Hallucination Rate | 9.5% | — |
| GPQA (HELM) | — | 59.4% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.5 (#66), gpt-oss-20b: 42.2 (#197)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1426 | 1268 |
| LMArena Chinese | 1499 | 1314 |
| LMArena German | 1447 | 1255 |
| LMArena Japanese | 1393 | 1244 |
| LMArena Korean | 1400 | 1236 |
| LMArena Russian | 1419 | 1278 |
| LMArena Spanish | 1436 | 1267 |
| LMArena French | 1459 | — |
Instruction Following GLM-4.6 leads
GLM-4.6: 74.3 (#98), gpt-oss-20b: 61.8 (#240)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1410 | 1236 |
| IFEval | — | 73.2% |
Long Context GLM-4.6 leads
GLM-4.6: 43.4 (#94), gpt-oss-20b: 37.9 (#209)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1422 | 1250 |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.1 (#90), gpt-oss-20b: 35.5 (#265)
| Benchmark | GLM-4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1440 | 1287 |
| LMArena Creative Writing | 1411 | 1201 |
| EQ-Bench Creative Writing | 1411 | 666 |
| LMArena Multi-Turn | 1427 | 1268 |
| WildBench | — | 73.7% |
Frequently asked questions
Is GLM-4.6 better than gpt-oss-20b?
GLM-4.6 is the stronger model overall, scoring 41.4 to 32.5 on the Noometry Index. gpt-oss-20b costs 28× less per token, which makes it the better buy when GLM-4.6's lead doesn't matter for your workload.
Which is cheaper, GLM-4.6 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-4.6 lists at $0.60 and $2.20.
Is GLM-4.6 or gpt-oss-20b better for coding?
GLM-4.6 scores higher on coding benchmarks: 40.1 versus 37.6 in the Noometry coding category.
Which has the bigger context window?
GLM-4.6 does, with 205K tokens against 131K.
How many benchmarks do GLM-4.6 and gpt-oss-20b share?
22 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and gpt-oss-20b has 34.