GLM 5.1 vs Step 5 Preview
Step 5 Preview leads GLM 5.1 on coding, 61.1 to 49.2 on the Vibe Coding Index, and Step 5 Preview costs 1.6x less per output token ($2.70 vs $4.40 per million). Updated October 22, 2026; 38 benchmarks have a result for both.
GLM 5.1 or Step 5 Preview: which to pick
Pick Step 5 Preview for coding: it scores 61.1 on the Vibe Coding Index against 49.2 for GLM 5.1.
Pick GLM 5.1 for work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
- Shared results
- 38
- GLM 5.1 only
- 51
- Step 5 Preview only
- 23
- Like-for-like categories
- 7 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
Step 5 Preview
Step 5 Preview scores 61.1 against 49.2.
3 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
Step 5 Preview
Step 5 Preview leads, 43.7 against 26.1.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only GLM 5.1 has a AA Agentic Index result, 23.9.
Directional only
Reasoning
GPQA Diamond
No call
Only GLM 5.1 has a GPQA Diamond result, 84.5%.
Directional only
Math
AIME
No call
Only GLM 5.1 has a AIME result, 91.9%.
Directional only
Knowledge
MMLU Pro
No call
Only GLM 5.1 has a MMLU Pro result, 86.9%.
Directional only
Web design
Design Arena: websites
No call
Only GLM 5.1 has a Design Arena: websites result, 1290.
Directional only
Long documents
Listed context window
Step 5 Preview
Step 5 Preview takes 1.05M tokens against 205K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Step 5 Preview
GLM 5.1 $0.0036, Step 5 Preview $0.0024 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GLM 5.1 has the number, $0.60.
Directional only
At a glance
- Coding index
- GLM 5.1 49.2 from 3 benchmarks, Step 5 Preview 61.1 from 4 benchmarks.
- Widest category
- Intelligence, 17.6 points apart on average over 1 shared benchmark.
- Widest single gap
- Vibe Code Bench (Vals AI): GLM 5.1 31.5% against Step 5 Preview 70.4%, 38.9 points apart on a 0 to 100 scale.
- Output price
- GLM 5.1 $4.40 against Step 5 Preview $2.70 per million output tokens, 1.6x apart.
- Context window
- GLM 5.1 takes 205K tokens, Step 5 Preview 1.05M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GLM 5.1 | Bars | Step 5 Preview | Ties |
|---|---|---|---|---|
| Coding | 2 | 6 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 2 | ||
| Reasoning | 2 | 2 | ||
| Math | 2 | 1 | ||
| Knowledge | 4 | 0 | ||
| General | 15 | 1 | ||
| All shared boards | 25 | 13 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index26
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 85.6% | Only Step 5 Preview measured |
| LMArena WebDev | 1508 | 1567 | Step 5 Preview |
| AA Coding Index | 55.8 | Not measured | Only GLM 5.1 measured |
| LiveCodeBench (Vals AI) | 81.4% | Not measured | Only GLM 5.1 measured |
| SWE-bench Verified (Vals AI) | 76.4% | Not measured | Only GLM 5.1 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 31.8% | Only Step 5 Preview measured |
| Vibe Code Bench (Vals AI) | 31.5% | 70.4% | Step 5 Preview |
| Code migration (Vals AI) | 25.8% | 39.0% | Step 5 Preview |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only GLM 5.1 measured |
| IOI (Vals AI) | Not measured | 38.1% | Only Step 5 Preview measured |
| Terminal-Bench 2.1 (Vals AI) | 56.9% | Not measured | Only GLM 5.1 measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 40.8% | Only Step 5 Preview measured |
| SciCode | 44.8% | 58.9% | Step 5 Preview |
| Terminal-Bench Hard (Artificial Analysis run) | 43.2% | Not measured | Only GLM 5.1 measured |
| WeirdML | 57.1% | Not measured | Only GLM 5.1 measured |
| ALE-Bench | 887.1 | Not measured | Only GLM 5.1 measured |
| LMArena Text: coding | 1485 | 1485 | GLM 5.1 |
| LMArena Text: software and IT | 1484 | 1473 | GLM 5.1 |
| LMArena WebDev: React | 1505 | 1583 | Step 5 Preview |
| LMArena WebDev: HTML | 1508 | 1556 | Step 5 Preview |
| LMArena WebDev: from an image | Not measured | 1550 | Only Step 5 Preview measured |
| Neither model has a result on the other 18 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| AA Intelligence Index | 26.1 | 43.7 | Step 5 Preview |
| Epoch Capabilities Index | 149.8 | Not measured | Only GLM 5.1 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| AA Agentic Index | 23.9 | Not measured | Only GLM 5.1 measured |
| Design Arena: agents building web apps | 1175 | Not measured | Only GLM 5.1 measured |
| Design Arena: agents building full-stack apps | 1167 | Not measured | Only GLM 5.1 measured |
| Design Arena: agents building mobile apps | 1161 | Not measured | Only GLM 5.1 measured |
| Design Arena: agents building Android apps | 1182 | Not measured | Only GLM 5.1 measured |
| Design Arena: agents building Godot games | 1053 | Not measured | Only GLM 5.1 measured |
| Design Arena: agents building games | 1164 | Not measured | Only GLM 5.1 measured |
| Design Arena: HTML slides | 1120 | Not measured | Only GLM 5.1 measured |
| Design Arena: PowerPoint slides | 1165 | Not measured | Only GLM 5.1 measured |
| Design Arena: spreadsheets | 1315 | Not measured | Only GLM 5.1 measured |
| LiveBench research | Not measured | 72.8% | Only Step 5 Preview measured |
| LiveBench browser | Not measured | 24.0% | Only Step 5 Preview measured |
| LiveBench long agentic | Not measured | 27.5% | Only Step 5 Preview measured |
| SREBench (Vals AI) | Not measured | 0.4% | Only Step 5 Preview measured |
| Finance Agent (Vals AI) | 44.8% | 50.7% | Step 5 Preview |
| Tax Agent Bench (Vals AI) | Not measured | 27.3% | Only Step 5 Preview measured |
| Legal Research Bench (Vals AI) | 27.9% | 40.4% | Step 5 Preview |
| Excel Modeling Benchmark (Vals AI) | Not measured | 57.1% | Only Step 5 Preview measured |
| BioMysteryBench (Vals AI) | Not measured | 67.4% | Only Step 5 Preview measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 2.9% | Only Step 5 Preview measured |
| Vals Science Index | Not measured | 29.6% | Only Step 5 Preview measured |
| τ²-Bench Telecom | 97.7% | Not measured | Only GLM 5.1 measured |
| APEX-Agents | 40.9% | Not measured | Only GLM 5.1 measured |
| Vending-Bench 2 | $5,634 | Not measured | Only GLM 5.1 measured |
| Neither model has a result on the other 17 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| GPQA Diamond | 84.5% | Not measured | Only GLM 5.1 measured |
| LiveBench reasoning | Not measured | 83.8% | Only Step 5 Preview measured |
| MysteryMechanism (Vals AI) | Not measured | 18.5% | Only Step 5 Preview measured |
| GPQA Diamond (Artificial Analysis run) | 86.8% | Not measured | Only GLM 5.1 measured |
| Humanity's Last Exam (Artificial Analysis run) | 30.1% | 46.5% | Step 5 Preview |
| GPQA Diamond (Epoch AI run) | 89.9% | Not measured | Only GLM 5.1 measured |
| SimpleBench | 55.1% | Not measured | Only GLM 5.1 measured |
| Chess puzzles | 20.0% | Not measured | Only GLM 5.1 measured |
| LMArena Text: hard prompts | 1473 | 1461 | GLM 5.1 |
| LMArena Text: expert | 1478 | 1485 | Step 5 Preview |
| LMArena Text: hard prompts in English | 1482 | 1454 | GLM 5.1 |
| LMArena Vision: diagrams | Not measured | 1308 | Only Step 5 Preview measured |
| LMArena Vision: homework | Not measured | 1265 | Only Step 5 Preview measured |
| Neither model has a result on the other 9 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index21
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| AIME | 91.9% | Not measured | Only GLM 5.1 measured |
| OTIS Mock AIME | 93.3% | Not measured | Only GLM 5.1 measured |
| FrontierMath, tiers 1 to 3 | 36.8% | Not measured | Only GLM 5.1 measured |
| ProofBench | 22.2% | 42.0% | Step 5 Preview |
| LMArena Text: math | 1473 | 1458 | GLM 5.1 |
| LMArena Text: mathematical work | 1470 | 1460 | GLM 5.1 |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| MMLU Pro | 86.9% | Not measured | Only GLM 5.1 measured |
| LegalBench (Vals AI) | 84.4% | Not measured | Only GLM 5.1 measured |
| CaseLaw (Vals AI) | 51.5% | Not measured | Only GLM 5.1 measured |
| CorpFin (Vals AI) | 64.5% | Not measured | Only GLM 5.1 measured |
| TaxEval (Vals AI) | 71.2% | Not measured | Only GLM 5.1 measured |
| MedCode (Vals AI) | 41.6% | Not measured | Only GLM 5.1 measured |
| SimpleQA Verified | 34.0% | Not measured | Only GLM 5.1 measured |
| LMArena Text: science | 1478 | 1454 | GLM 5.1 |
| LMArena Text: medicine and healthcare | 1474 | 1439 | GLM 5.1 |
| LMArena Text: legal and government | 1465 | 1437 | GLM 5.1 |
| LMArena Text: business and finance | 1452 | 1435 | GLM 5.1 |
| Neither model has a result on the other 7 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| Design Arena: websites | 1290 | Not measured | Only GLM 5.1 measured |
| Design Arena: UI components | 1292 | Not measured | Only GLM 5.1 measured |
| Design Arena: game dev | 1264 | Not measured | Only GLM 5.1 measured |
| Design Arena: data viz | 1366 | Not measured | Only GLM 5.1 measured |
| Design Arena: 3D | 1336 | Not measured | Only GLM 5.1 measured |
| Design Arena: all code categories | 1268 | Not measured | Only GLM 5.1 measured |
| Design Arena: SVG | 1217 | Not measured | Only GLM 5.1 measured |
| Design Arena: ASCII art | 1141 | Not measured | Only GLM 5.1 measured |
| LiveBench design | Not measured | 76.9% | Only Step 5 Preview measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index151
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 63.5% | Only Step 5 Preview measured |
| Public Benefits Bench (Vals AI) | 61.8% | 58.1% | GLM 5.1 |
| MedScribe (Vals AI) | 72.3% | Not measured | Only GLM 5.1 measured |
| IFBench | 76.3% | Not measured | Only GLM 5.1 measured |
| Long context reasoning (AA-LCR) | 73.7% | 88.3% | Step 5 Preview |
| ExploitBench | 18.1% | Not measured | Only GLM 5.1 measured |
| LMArena Text | 1463 | 1440 | GLM 5.1 |
| LMArena Text: instruction following | 1453 | 1438 | GLM 5.1 |
| LMArena Text: creative writing | 1453 | 1409 | GLM 5.1 |
| LMArena Vision | Not measured | 1265 | Only Step 5 Preview measured |
| LMArena Text: English | 1476 | 1447 | GLM 5.1 |
| LMArena Text: non-English | 1448 | 1428 | GLM 5.1 |
| LMArena Text: Chinese | 1515 | 1500 | GLM 5.1 |
| LMArena Text: German | 1467 | Not measured | Only GLM 5.1 measured |
| LMArena Text: Japanese | 1435 | 1383 | GLM 5.1 |
| LMArena Text: Korean | 1424 | Not measured | Only GLM 5.1 measured |
| LMArena Text: French | 1470 | 1438 | GLM 5.1 |
| LMArena Text: Spanish | 1468 | 1441 | GLM 5.1 |
| LMArena Text: Russian | 1454 | 1433 | GLM 5.1 |
| LMArena Text: Polish | 1449 | Not measured | Only GLM 5.1 measured |
| LMArena Text: multi-turn | 1473 | 1437 | GLM 5.1 |
| LMArena Text: longer queries | 1467 | 1449 | GLM 5.1 |
| LMArena Text: writing and language | 1454 | 1425 | GLM 5.1 |
| LMArena Text: entertainment and media | 1442 | 1412 | GLM 5.1 |
| LMArena Vision: English | Not measured | 1290 | Only Step 5 Preview measured |
| LMArena Vision: reading text | Not measured | 1282 | Only Step 5 Preview measured |
| Neither model has a result on the other 20 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GLM 5.1 | Step 5 Preview | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Step 5 Preview costs less
Repository review
50K input + 3K output tokens
Step 5 Preview costs less
Agent loop
200K input + 10K output tokens
Step 5 Preview costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Z.ai GLM 5.1 | StepFun Step 5 Preview |
|---|---|---|
| Maker | Z.ai | StepFun |
| Weights | Open weights | Open weights |
| Released | April 7, 2026 | October 8, 2026 |
| Context window | 205K | 1.05M |
| Input $/Mtok | $1.40 | $1.00 |
| Output $/Mtok | $4.40 | $2.70 |
| Coding index | 49.2 | 61.1 |
| Independent measurements behind it | 3 | 4 |
| Cost per solved task | $0.60 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GLM 5.1 better than Step 5 Preview for coding?
Pick Step 5 Preview for coding: it scores 61.1 on the Vibe Coding Index against 49.2 for GLM 5.1. Pick GLM 5.1 for work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
Which is cheaper, GLM 5.1 or Step 5 Preview?
GLM 5.1 costs $1.40 per million input tokens and $4.40 per million output tokens; Step 5 Preview costs $1.00 and $2.70. Cost per solved task is known for GLM 5.1 only.
Which has the bigger context window, GLM 5.1 or Step 5 Preview?
Step 5 Preview, with 1.05M tokens against 205K.
How many benchmarks is the GLM 5.1 vs Step 5 Preview comparison based on?
38 benchmarks have a result for both models as of October 22, 2026. GLM 5.1's coding index rests on 3 independent benchmarks and Step 5 Preview's on 4. Where only one model has a result, the page says so instead of naming a winner.