Model comparison
GLM-4.5 vs Step 3
GLM-4.5 is the stronger model overall, scoring 42.0 to 40.5 on the Noometry Index.
Last verified . 16 shared benchmarks.
StepFunSummary
- They share 16 benchmarks with published results for both. GLM-4.5 scores higher in 6 categories and Step 3 in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where GLM-4.5 leads 52.8 to 46.3.
Side by side
| GLM-4.5 | Step 3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | StepFun |
| Noometry Index | 42.0 | 40.5 |
| Released | 2025-07-27 | — |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 98K | — |
| Input $ / M tokens | $0.60 | — |
| Output $ / M tokens | $2.20 | — |
| Results tracked | 27 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.5 leads
GLM-4.5: 41.4 (#125), Step 3: 40.1 (#147)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Coding | 1434 | 1367 |
| SWE-bench Verified (bash only) | 54.2% | — |
| WeirdML | 40.6% | — |
| ALE-Bench | 344.82 | — |
| AlgoTune | 1.52 | — |
Reasoning Too close to call
GLM-4.5: 28.6 (#100), Step 3: 28.4 (#105)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 57.9% | 62.3% |
| LMArena Hard Prompts | 1429 | 1355 |
Math GLM-4.5 leads
GLM-4.5: 39.0 (#116), Step 3: 37.6 (#148)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Math | 1427 | 1366 |
Knowledge Too close to call
GLM-4.5: 35.9 (#179), Step 3: 36.8 (#164)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Expert | 1433 | 1333 |
| Humanity's Last Exam | 8.3% | — |
| Confabulations | 11.3% | — |
Multimodal Not comparable
GLM-4.5: —, Step 3: 35.5 (#86)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual GLM-4.5 leads
GLM-4.5: 52.8 (#77), Step 3: 46.3 (#159)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Non-English | 1417 | 1327 |
| LMArena Chinese | 1465 | 1397 |
| LMArena German | 1407 | 1371 |
| LMArena Korean | 1380 | 1269 |
| LMArena Russian | 1414 | 1331 |
| LMArena Spanish | 1454 | 1371 |
| LMArena French | 1418 | — |
| LMArena Japanese | 1415 | — |
Instruction Following GLM-4.5 leads
GLM-4.5: 74.1 (#104), Step 3: 70.4 (#164)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1332 |
Long Context Step 3 leads
GLM-4.5: 38.2 (#201), Step 3: 40.3 (#157)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1412 | 1326 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference GLM-4.5 leads
GLM-4.5: 57.5 (#127), Step 3: 54.3 (#151)
| Benchmark | GLM-4.5 | Step 3 |
|---|---|---|
| LMArena Text | 1430 | 1350 |
| LMArena Creative Writing | 1395 | 1321 |
| LMArena Multi-Turn | 1415 | 1341 |
| Short-Story Creative Writing | 73.4% | — |
| EQ-Bench Creative Writing | 1343 | — |
Frequently asked questions
Is GLM-4.5 better than Step 3?
GLM-4.5 is the stronger model overall, scoring 42.0 to 40.5 on the Noometry Index.
Is GLM-4.5 or Step 3 better for coding?
GLM-4.5 scores higher on coding benchmarks: 41.4 versus 40.1 in the Noometry coding category.
How many benchmarks do GLM-4.5 and Step 3 share?
16 benchmarks have published results for both models. GLM-4.5 has 27 scored results on Noometry and Step 3 has 17.