GLM 5.3 vs Inkling
Updated October 21, 2026. 93 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick GLM 5.3 for coding: it scores 73.1 on the Vibe Coding Index against 49.7 for Inkling.
Pick Inkling when what matters is LiveCodeBench (Vals AI) or LiveBench browser, where it leads.
- Shared results
- 93
- GLM 5.3 only
- 7
- Inkling only
- 9
- Like-for-like categories
- 8 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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, four public coding benchmarks
GLM 5.3
GLM 5.3 scores 73.1 against 49.7.
2 benchmarks behind each score
Intelligence
AA Intelligence Index
GLM 5.3
GLM 5.3 leads, 44.8 against 25.0.
Same board, both measured
Agentic work
AA Agentic Index
GLM 5.3
GLM 5.3 leads, 53.1 against 22.5.
Same board, both measured
Reasoning
GPQA Diamond
GLM 5.3
GLM 5.3 leads, 88.1% against 87.1%.
Same board, both measured
Knowledge
MMLU Pro
GLM 5.3
GLM 5.3 leads, 86.8% against 86.3%.
Same board, both measured
Web design
Design Arena: websites
GLM 5.3
GLM 5.3 leads, 1295 against 1224.
Same board, both measured
General
Vals Index
GLM 5.3
GLM 5.3 leads, 59.7% against 29.0%.
Same board, both measured
Long documents
Listed context window
GLM 5.3
GLM 5.3 takes 1.05M tokens against 524K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GLM 5.3
GLM 5.3 $0.0008, Inkling $0.0030 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GLM 5.3
GLM 5.3 $0.36, Inkling $0.76.
Published attempt cost and pass rate
At a glance
- Coding index
- GLM 5.3 73.1 from 2 benchmarks, Inkling 49.7 from 2 benchmarks.
- Widest category
- Coding, 23.3 points apart on average over 23 shared benchmarks.
- Widest single gap
- Vibe Code Bench (Vals AI): GLM 5.3 78.1% against Inkling 19.2%, 58.9 points apart on a 0 to 100 scale.
- Output price
- GLM 5.3 $1.40 against Inkling $4.05 per million output tokens, 2.9x apart.
- Context window
- GLM 5.3 takes 1.05M tokens, Inkling 524K.
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 5.3 | Bars | Inkling | Ties |
|---|---|---|---|---|
| Coding | 22 | 1 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 10 | 1 | ||
| Reasoning | 9 | 0 | ||
| Math | 6 | 0 | ||
| Knowledge | 8 | 1 | ||
| Web design | 9 | 0 | ||
| General | 23 | 1 | ||
| All shared boards | 89 | 4 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index221
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| LiveBench coding | 82.8% | 64.0% | GLM 5.3 |
| LMArena WebDev | 1621 | 1414 | GLM 5.3 |
| AA Coding Index | 74.8 | 52.1 | GLM 5.3 |
| LiveCodeBench (Vals AI) | 80.5% | 85.5% | Inkling |
| SWE-bench Verified (Vals AI) | 95.4% | 77.6% | GLM 5.3 |
| Terminal-Bench 4 (Vals AI) | 38.9% | 0.5% | GLM 5.3 |
| Vibe Code Bench (Vals AI) | 78.1% | 19.2% | GLM 5.3 |
| Code migration (Vals AI) | 44.2% | 11.8% | GLM 5.3 |
| ProgramBench (Vals AI) | 1.5% | 0.0% | GLM 5.3 |
| IOI (Vals AI) | 68.4% | 14.9% | GLM 5.3 |
| Terminal-Bench 2.1 (Vals AI) | 71.5% | 47.6% | GLM 5.3 |
| Vibe Code Bench 1-100 (Vals AI) | 20.0% | 7.3% | GLM 5.3 |
| SkillsBench (Vals AI) | 47.5% | 26.1% | GLM 5.3 |
| CyberBench (Vals AI) | 72.1% | 66.8% | GLM 5.3 |
| SciCode | 59.0% | 47.0% | GLM 5.3 |
| FrontierCode | 40.1% | 14.0% | GLM 5.3 |
| FrontierSWE | 30.2% | 4.1% | GLM 5.3 |
| WeirdML | 75.4% | 32.3% | GLM 5.3 |
| ALE-Bench | 1317.4 | 946.0 | GLM 5.3 |
| LMArena Text: coding | 1497 | 1466 | GLM 5.3 |
| LMArena Text: software and IT | 1493 | 1465 | GLM 5.3 |
| LMArena WebDev: React | 1641 | 1404 | GLM 5.3 |
| LMArena WebDev: HTML | 1595 | 1399 | GLM 5.3 |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| AA Intelligence Index | 44.8 | 25.0 | GLM 5.3 |
| Epoch Capabilities Index | 156.0 | 148.4 | GLM 5.3 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index101
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| AA Agentic Index | 53.1 | 22.5 | GLM 5.3 |
| Design Arena: agents building web apps | Not measured | 1196 | Only Inkling measured |
| Design Arena: agents building mobile apps | 1194 | Not measured | Only GLM 5.3 measured |
| Design Arena: HTML slides | 1189 | Not measured | Only GLM 5.3 measured |
| Design Arena: PowerPoint slides | 1255 | Not measured | Only GLM 5.3 measured |
| LiveBench research | 73.2% | 59.9% | GLM 5.3 |
| LiveBench browser | 0.0% | 16.0% | Inkling |
| LiveBench long agentic | 35.9% | 14.1% | GLM 5.3 |
| SREBench (Vals AI) | 9.2% | 0.0% | GLM 5.3 |
| Finance Agent (Vals AI) | 55.8% | 46.6% | GLM 5.3 |
| Tax Agent Bench (Vals AI) | 39.7% | 15.3% | GLM 5.3 |
| Legal Research Bench (Vals AI) | 49.0% | 28.4% | GLM 5.3 |
| Harvey's Legal Agent Benchmark (Vals AI) | 14.6% | 3.8% | GLM 5.3 |
| Excel Modeling Benchmark (Vals AI) | 56.3% | 38.4% | GLM 5.3 |
| Terminal-Bench-Science (Vals AI) | 5.7% | Not measured | Only GLM 5.3 measured |
| APEX-Agents | 56.6% | 33.8% | GLM 5.3 |
| Vending-Bench 2 | $8,164 | Not measured | Only GLM 5.3 measured |
| Neither model has a result on the other 22 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index90
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| GPQA Diamond | 88.1% | 87.1% | GLM 5.3 |
| LiveBench reasoning | 85.8% | 78.3% | GLM 5.3 |
| MysteryMechanism (Vals AI) | 23.0% | Not measured | Only GLM 5.3 measured |
| MMMU Pro (Vals AI) | Not measured | 78.5% | Only Inkling measured |
| GPQA Diamond (Artificial Analysis run) | 91.7% | 87.2% | GLM 5.3 |
| Humanity's Last Exam (Artificial Analysis run) | 42.3% | 31.9% | GLM 5.3 |
| GPQA Diamond (Epoch AI run) | 90.9% | 88.3% | GLM 5.3 |
| ARC-AGI | Not measured | 79.5% | Only Inkling measured |
| ARC-AGI-2 | Not measured | 36.5% | Only Inkling measured |
| SimpleBench | Not measured | 50.0% | Only Inkling measured |
| Chess puzzles | 37.0% | 21.0% | GLM 5.3 |
| Mystery game puzzles | 33.0% | Not measured | Only GLM 5.3 measured |
| LMArena Text: hard prompts | 1490 | 1451 | GLM 5.3 |
| LMArena Text: expert | 1514 | 1466 | GLM 5.3 |
| LMArena Text: hard prompts in English | 1496 | 1448 | GLM 5.3 |
| Neither model has a result on the other 6 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| OTIS Mock AIME | 91.1% | 88.9% | GLM 5.3 |
| FrontierMath, tiers 1 to 3 | 68.8% | 33.3% | GLM 5.3 |
| FrontierMath, tier 4 | 29.3% | 4.9% | GLM 5.3 |
| ProofBench | 49.0% | 0.0% | GLM 5.3 |
| LMArena Text: math | 1495 | 1479 | GLM 5.3 |
| LMArena Text: mathematical work | 1497 | 1460 | GLM 5.3 |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index81
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| MMLU Pro | 86.8% | 86.3% | GLM 5.3 |
| LegalBench (Vals AI) | 84.8% | 83.0% | GLM 5.3 |
| CorpFin (Vals AI) | Not measured | 68.6% | Only Inkling measured |
| TaxEval (Vals AI) | 72.4% | 75.3% | Inkling |
| MortgageTax (Vals AI) | Not measured | 64.0% | Only Inkling measured |
| MedCode (Vals AI) | 42.9% | 41.2% | GLM 5.3 |
| SimpleQA Verified | 41.0% | 40.3% | GLM 5.3 |
| LMArena Text: science | 1498 | 1455 | GLM 5.3 |
| LMArena Text: medicine and healthcare | 1481 | 1447 | GLM 5.3 |
| LMArena Text: legal and government | 1473 | 1448 | GLM 5.3 |
| LMArena Text: business and finance | 1459 | 1438 | GLM 5.3 |
| Neither model has a result on the other 6 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index90
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| Design Arena: websites | 1295 | 1224 | GLM 5.3 |
| Design Arena: UI components | 1314 | 1200 | GLM 5.3 |
| Design Arena: game dev | 1311 | 1166 | GLM 5.3 |
| Design Arena: data viz | 1260 | 1160 | GLM 5.3 |
| Design Arena: 3D | 1332 | 1161 | GLM 5.3 |
| Design Arena: all code categories | 1304 | 1206 | GLM 5.3 |
| Design Arena: SVG | 1292 | 1109 | GLM 5.3 |
| Design Arena: ASCII art | 1213 | 1089 | GLM 5.3 |
| LiveBench design | 77.3% | 58.3% | GLM 5.3 |
GeneralLike-for-likePublic boards, shown as published, kept out of the index231
| Benchmark | GLM 5.3 | Inkling | Leader |
|---|---|---|---|
| Vals Index | 59.7% | 29.0% | GLM 5.3 |
| LiveBench data analysis | 70.2% | 72.8% | Inkling |
| Public Benefits Bench (Vals AI) | 68.5% | 58.5% | GLM 5.3 |
| SAGE (Vals AI) | Not measured | 36.5% | Only Inkling measured |
| MedScribe (Vals AI) | 88.8% | 85.4% | GLM 5.3 |
| Vals Multimodal Index | Not measured | 49.4% | Only Inkling measured |
| Long context reasoning (AA-LCR) | 79.7% | 77.3% | GLM 5.3 |
| DTBench | 87.7% | 87.5% | GLM 5.3 |
| LMCA | 55.5% | 37.6% | GLM 5.3 |
| LMArena Text | 1474 | 1442 | GLM 5.3 |
| LMArena Text: instruction following | 1479 | 1425 | GLM 5.3 |
| LMArena Text: creative writing | 1458 | 1390 | GLM 5.3 |
| LMArena Text: English | 1482 | 1442 | GLM 5.3 |
| LMArena Text: non-English | 1459 | 1434 | GLM 5.3 |
| LMArena Text: Chinese | 1532 | 1493 | GLM 5.3 |
| LMArena Text: German | 1488 | 1445 | GLM 5.3 |
| LMArena Text: Japanese | 1460 | 1429 | GLM 5.3 |
| LMArena Text: Korean | 1460 | 1403 | GLM 5.3 |
| LMArena Text: French | 1503 | 1455 | GLM 5.3 |
| LMArena Text: Spanish | 1462 | 1449 | GLM 5.3 |
| LMArena Text: Russian | 1469 | 1430 | GLM 5.3 |
| LMArena Text: Polish | 1466 | 1450 | GLM 5.3 |
| LMArena Text: multi-turn | 1472 | 1437 | GLM 5.3 |
| LMArena Text: longer queries | 1483 | 1435 | GLM 5.3 |
| LMArena Text: writing and language | 1469 | 1404 | GLM 5.3 |
| LMArena Text: entertainment and media | 1447 | 1394 | GLM 5.3 |
| 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 5.3 | Inkling | 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
GLM 5.3 costs less
Repository review
50K input + 3K output tokens
GLM 5.3 costs less
Agent loop
200K input + 10K output tokens
GLM 5.3 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 5.3 | Thinking Machines Inkling |
|---|---|---|
| Maker | Z.ai | Thinking Machines |
| Weights | Open weights | Open weights |
| Released | August 18, 2026 | July 17, 2026 |
| Context window | 1.05M | 524K |
| Input $/Mtok | $0.079 | $1.00 |
| Output $/Mtok | $1.40 | $4.05 |
| Coding index | 73.1 | 49.7 |
| Independent measurements behind it | 2 | 2 |
| Cost per solved task | $0.36 | $0.76 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GLM 5.3 better than Inkling for coding?
Pick GLM 5.3 for coding: it scores 73.1 on the Vibe Coding Index against 49.7 for Inkling. Pick Inkling when what matters is LiveCodeBench (Vals AI) or LiveBench browser, where it leads.
Which is cheaper, GLM 5.3 or Inkling?
GLM 5.3 costs $0.079 per million input tokens and $1.40 per million output tokens; Inkling costs $1.00 and $4.05. Per solved SWE-bench Verified task, GLM 5.3 costs $0.36 and Inkling $0.76.
Which has the bigger context window, GLM 5.3 or Inkling?
GLM 5.3, with 1.05M tokens against 524K.
How many benchmarks is the GLM 5.3 vs Inkling comparison based on?
93 benchmarks have a result for both models as of October 21, 2026. GLM 5.3's coding index rests on 2 independent benchmarks and Inkling's on 2. Where only one model has a result, the page says so instead of naming a winner.