Model comparison
GLM-5.1 vs gpt-oss-20b
GLM-5.1 is the stronger model overall, scoring 47.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 60× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. GLM-5.1 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-5.1 leads 66.9 to 35.5.
- The biggest single-benchmark swing is GPQA Diamond: 89.9% for GLM-5.1 and 60.8% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $1.40 / $4.40 for GLM-5.1.
- GLM-5.1 accepts more context: 200K tokens versus 131K.
Side by side
| GLM-5.1 | gpt-oss-20b | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 47.8 | 32.5 |
| Released | 2026-04-07 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 200K | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $1.40 | $0.018 |
| Output $ / M tokens | $4.40 | $0.09 |
| Results tracked | 41 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
GLM-5.1: 48.7 (#55), gpt-oss-20b: 37.6 (#192)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| SciCode | 43.8% | 34.4% |
| WeirdML | 57.1% | 40.9% |
| LMArena Coding | 1485 | 1306 |
| ALE-Bench | 887.1 | 566.05 |
| SWE-bench Verified | 74.2% | — |
| LMArena WebDev | 1508 | — |
Agentic & Tool Use GLM-5.1 leads
GLM-5.1: 24.9 (#113), gpt-oss-20b: 9.3 (#154)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| APEX-Agents | 40.9% | — |
| ExploitBench | 18.1% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 5,634 | — |
Reasoning GLM-5.1 leads
GLM-5.1: 39.1 (#60), gpt-oss-20b: 19.3 (#261)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| CritPt | 4.6% | 1.4% |
| Chess Puzzles | 19% | 4% |
| LMArena Hard Prompts | 1472 | 1274 |
| Epoch Capabilities Index | 149.84 | 137.82 |
| SimpleBench | 55.1% | — |
| Kagi LLM Benchmark | — | 53.2% |
| NYT Connections (extended) | 77.7% | — |
| Thematic Generalization | 69.8% | — |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
Math GLM-5.1 leads
GLM-5.1: 49.7 (#60), gpt-oss-20b: 39.4 (#103)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.3% | 65.3% |
| LMArena Math | 1473 | 1317 |
| FrontierMath (Tiers 1-3) | 36.8% | — |
| MathArena Final-Answer Competitions | 67.1% | — |
| ProofBench | 22.2% | — |
| Omni-MATH | — | 56.5% |
| FrontierMath (Feb 2025 set) | 33.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GLM-5.1 leads
GLM-5.1: 54.9 (#50), gpt-oss-20b: 34.6 (#195)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 89.9% | 60.8% |
| LMArena Expert | 1476 | 1258 |
| SimpleQA Verified | 34% | — |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |
Multilingual GLM-5.1 leads
GLM-5.1: 55.0 (#36), gpt-oss-20b: 42.2 (#197)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1447 | 1268 |
| LMArena Chinese | 1515 | 1314 |
| LMArena German | 1465 | 1255 |
| LMArena Japanese | 1434 | 1244 |
| LMArena Korean | 1418 | 1236 |
| LMArena Russian | 1454 | 1278 |
| LMArena Spanish | 1469 | 1267 |
| LMArena French | 1474 | — |
Instruction Following GLM-5.1 leads
GLM-5.1: 76.3 (#42), gpt-oss-20b: 61.8 (#240)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1451 | 1236 |
| IFEval | — | 73.2% |
Long Context GLM-5.1 leads
GLM-5.1: 44.9 (#53), gpt-oss-20b: 37.9 (#209)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1466 | 1250 |
Writing & Preference GLM-5.1 leads
GLM-5.1: 66.9 (#31), gpt-oss-20b: 35.5 (#265)
| Benchmark | GLM-5.1 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1461 | 1287 |
| LMArena Creative Writing | 1453 | 1201 |
| EQ-Bench Creative Writing | 1592 | 666 |
| LMArena Multi-Turn | 1472 | 1268 |
| WildBench | — | 73.7% |
Frequently asked questions
Is GLM-5.1 better than gpt-oss-20b?
GLM-5.1 is the stronger model overall, scoring 47.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 60× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Which is cheaper, GLM-5.1 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-5.1 lists at $1.40 and $4.40.
Is GLM-5.1 or gpt-oss-20b better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.7 versus 37.6 in the Noometry coding category.
Which has the bigger context window?
GLM-5.1 does, with 200K tokens against 131K.
How many benchmarks do GLM-5.1 and gpt-oss-20b share?
25 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and gpt-oss-20b has 34.