Model comparison
GLM-4.7-Flash vs gpt-oss-20b
GLM-4.7-Flash is the stronger model overall, scoring 38.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 4.0× less per token, which makes it the better buy when GLM-4.7-Flash's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. GLM-4.7-Flash scores higher in 7 categories and gpt-oss-20b in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.7-Flash leads 47.4 to 35.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 58.3% for GLM-4.7-Flash and 65.3% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.06 / $0.40 for GLM-4.7-Flash.
- GLM-4.7-Flash accepts more context: 200K tokens versus 131K.
Side by side
| GLM-4.7-Flash | gpt-oss-20b | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 38.8 | 32.5 |
| Released | 2026-01-19 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 200K | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $0.06 | $0.018 |
| Output $ / M tokens | $0.40 | $0.09 |
| Results tracked | 21 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.7-Flash leads
GLM-4.7-Flash: 40.6 (#135), gpt-oss-20b: 37.6 (#192)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Coding | 1383 | 1306 |
| SciCode | — | 34.4% |
| WeirdML | — | 40.9% |
| ALE-Bench | — | 566.05 |
Agentic & Tool Use Not comparable
GLM-4.7-Flash: —, gpt-oss-20b: 9.3 (#154)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
Reasoning GLM-4.7-Flash leads
GLM-4.7-Flash: 20.9 (#229), gpt-oss-20b: 19.3 (#261)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| Chess Puzzles | 0% | 4% |
| LMArena Hard Prompts | 1356 | 1274 |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| Epoch Capabilities Index | — | 137.82 |
Math gpt-oss-20b leads
GLM-4.7-Flash: 36.1 (#173), gpt-oss-20b: 39.4 (#103)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 58.3% | 65.3% |
| LMArena Math | 1355 | 1317 |
| Omni-MATH | — | 56.5% |
Knowledge Too close to call
GLM-4.7-Flash: 35.5 (#184), gpt-oss-20b: 34.6 (#195)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 60.5% | 60.8% |
| LMArena Expert | 1357 | 1258 |
| MMLU-Pro | — | 74% |
| Vectara Hallucination Rate | 9.3% | — |
| GPQA (HELM) | — | 59.4% |
Multilingual GLM-4.7-Flash leads
GLM-4.7-Flash: 46.5 (#158), gpt-oss-20b: 42.2 (#197)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1330 | 1268 |
| LMArena Chinese | 1403 | 1314 |
| LMArena German | 1337 | 1255 |
| LMArena Korean | 1283 | 1236 |
| LMArena Russian | 1332 | 1278 |
| LMArena Spanish | 1350 | 1267 |
| LMArena French | 1332 | — |
| LMArena Japanese | — | 1244 |
Instruction Following GLM-4.7-Flash leads
GLM-4.7-Flash: 70.1 (#167), gpt-oss-20b: 61.8 (#240)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1327 | 1236 |
| IFEval | — | 73.2% |
Long Context GLM-4.7-Flash leads
GLM-4.7-Flash: 40.9 (#148), gpt-oss-20b: 37.9 (#209)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1345 | 1250 |
Writing & Preference GLM-4.7-Flash leads
GLM-4.7-Flash: 47.4 (#210), gpt-oss-20b: 35.5 (#265)
| Benchmark | GLM-4.7-Flash | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1351 | 1287 |
| LMArena Creative Writing | 1297 | 1201 |
| EQ-Bench Creative Writing | 1125 | 666 |
| LMArena Multi-Turn | 1342 | 1268 |
| WildBench | — | 73.7% |
Frequently asked questions
Is GLM-4.7-Flash better than gpt-oss-20b?
GLM-4.7-Flash is the stronger model overall, scoring 38.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 4.0× less per token, which makes it the better buy when GLM-4.7-Flash's lead doesn't matter for your workload.
Which is cheaper, GLM-4.7-Flash 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-4.7-Flash lists at $0.06 and $0.40.
Is GLM-4.7-Flash or gpt-oss-20b better for coding?
GLM-4.7-Flash scores higher on coding benchmarks: 40.6 versus 37.6 in the Noometry coding category.
Which has the bigger context window?
GLM-4.7-Flash does, with 200K tokens against 131K.
How many benchmarks do GLM-4.7-Flash and gpt-oss-20b share?
19 benchmarks have published results for both models. GLM-4.7-Flash has 21 scored results on Noometry and gpt-oss-20b has 34.