Model comparison
GLM-5 vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 46.1 on the Noometry Index.
Last verified . 14 shared benchmarks.
StepFunSummary
- They share 14 benchmarks with published results for both. GLM-5 scores higher in 4 categories and Step 5 Preview in 4 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 5 Preview leads 40.0 to 27.6.
- Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $1 / $3.20 for GLM-5.
- Step 5 Preview accepts more context: 1.02M tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | Step 5 Preview | |
|---|---|---|
| Provider | Z.ai (Zhipu) | StepFun |
| Noometry Index | 46.1 | 47.9 |
| Released | 2026-02-11 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | 205K | 1.02M |
| Max output | 131K | 66K |
| Input $ / M tokens | $1 | $1 |
| Output $ / M tokens | $3.20 | $2.70 |
| Results tracked | 45 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
GLM-5: 49.0 (#52), Step 5 Preview: 51.8 (#36)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena WebDev | 1434 | 1564 |
| LMArena Coding | 1461 | 1480 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 58.9% |
| WeirdML | 48.2% | — |
| ALE-Bench | 765.62 | — |
Agentic & Tool Use Not comparable
GLM-5: 31.1 (#71), Step 5 Preview: —
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| 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 Step 5 Preview leads
GLM-5: 27.6 (#116), Step 5 Preview: 40.0 (#57)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1452 | 1465 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| CritPt | — | 20.9% |
| Chess Puzzles | 10% | — |
| Epoch Capabilities Index | 145.83 | — |
| ForecastBench | 61 | — |
Math Too close to call
GLM-5: 46.4 (#71), Step 5 Preview: 46.1 (#72)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Math | 1440 | 1470 |
| MathArena Final-Answer Competitions | 65.7% | — |
| OTIS Mock AIME 2024-2025 | 80% | — |
| ProofBench | — | 42% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), Step 5 Preview: 41.2 (#112)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1454 | 1470 |
| GPQA Diamond | 87.8% | — |
| Vectara Hallucination Rate | 10.1% | — |
Multimodal Not comparable
GLM-5: —, Step 5 Preview: 41.0 (#41)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Too close to call
GLM-5: 53.7 (#58), Step 5 Preview: 53.8 (#54)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1430 | 1432 |
| LMArena Chinese | 1511 | 1519 |
| LMArena Russian | 1436 | 1436 |
| LMArena Spanish | 1454 | 1439 |
| LMArena French | 1455 | — |
| LMArena German | 1445 | — |
| LMArena Japanese | 1416 | — |
| LMArena Korean | 1423 | — |
Instruction Following Too close to call
GLM-5: 75.2 (#67), Step 5 Preview: 76.0 (#50)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1428 | 1444 |
Long Context Too close to call
GLM-5: 44.7 (#60), Step 5 Preview: 44.5 (#64)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1446 | 1455 |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), Step 5 Preview: 62.8 (#71)
| Benchmark | GLM-5 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1446 | 1442 |
| LMArena Creative Writing | 1439 | 1410 |
| LMArena Multi-Turn | 1456 | 1447 |
| EQ-Bench Creative Writing | 1601 | — |
Frequently asked questions
Is GLM-5 better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 46.1 on the Noometry Index.
Which is cheaper, GLM-5 or Step 5 Preview?
Step 5 Preview is cheaper. It lists at $1 per million input tokens and $2.70 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-5 or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 49.0 in the Noometry coding category.
Which has the bigger context window?
Step 5 Preview does, with 1.02M tokens against 205K.
How many benchmarks do GLM-5 and Step 5 Preview share?
14 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Step 5 Preview has 18.