Model comparison
GLM-4.5-Air vs Inkling
Inkling is the stronger model overall, scoring 44.1 to 38.9 on the Noometry Index. GLM-4.5-Air costs 6.1× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Summary
- They share 17 benchmarks with published results for both. GLM-4.5-Air scores higher in 1 category and Inkling in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.1 to 35.0.
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- GLM-4.5-Air accepts more context: 131K tokens versus 66K.
Side by side
| GLM-4.5-Air | Inkling | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Thinking Machines Lab |
| Noometry Index | 38.9 | 44.1 |
| Released | 2025-07-20 | 2026-07-15 |
| Weights | Open | Open |
| Context window | 131K | 66K |
| Max output | 98K | 66K |
| Input $ / M tokens | $0.20 | $1.87 |
| Output $ / M tokens | $1.10 | $4.68 |
| Results tracked | 27 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling leads
GLM-4.5-Air: 33.3 (#259), Inkling: 34.5 (#234)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Coding | 1397 | 1464 |
| FrontierCode | — | 14% |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| GSO | 2.9% | — |
| WeirdML | — | 32.3% |
| ALE-Bench | — | 946 |
Agentic & Tool Use Not comparable
GLM-4.5-Air: —, Inkling: 29.6 (#85)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |
Reasoning Inkling leads
GLM-4.5-Air: 24.1 (#166), Inkling: 40.4 (#56)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Hard Prompts | 1379 | 1451 |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| Kagi LLM Benchmark | 43% | — |
| ARC-AGI-1 | — | 79.5% |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| Epoch Capabilities Index | — | 148.54 |
| ForecastBench | 59.2 | — |
Math GLM-4.5-Air leads
GLM-4.5-Air: 36.2 (#170), Inkling: 31.3 (#225)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Math | 1396 | 1479 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |
| Omni-MATH | 39.1% | — |
Knowledge Inkling leads
GLM-4.5-Air: 35.0 (#191), Inkling: 55.1 (#49)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Expert | 1370 | 1465 |
| GPQA Diamond | — | 88.3% |
| Humanity's Last Exam | 8.1% | — |
| SimpleQA Verified | — | 40.3% |
| MMLU-Pro | 76.2% | — |
| Vectara Hallucination Rate | 9.3% | — |
| GPQA (HELM) | 59.4% | — |
Multilingual Inkling leads
GLM-4.5-Air: 49.1 (#135), Inkling: 54.0 (#52)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Non-English | 1366 | 1434 |
| LMArena Chinese | 1426 | 1490 |
| LMArena French | 1399 | 1458 |
| LMArena German | 1377 | 1446 |
| LMArena Japanese | 1348 | 1429 |
| LMArena Korean | 1308 | 1404 |
| LMArena Russian | 1373 | 1429 |
| LMArena Spanish | 1386 | 1448 |
Instruction Following Inkling leads
GLM-4.5-Air: 69.6 (#171), Inkling: 75.1 (#71)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Instruction Following | 1354 | 1426 |
| IFEval | 81.2% | — |
Long Context Inkling leads
GLM-4.5-Air: 41.6 (#135), Inkling: 43.8 (#86)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Longer Query | 1366 | 1434 |
Writing & Preference Inkling leads
GLM-4.5-Air: 55.9 (#139), Inkling: 65.2 (#51)
| Benchmark | GLM-4.5-Air | Inkling |
|---|---|---|
| LMArena Text | 1384 | 1441 |
| LMArena Creative Writing | 1343 | 1387 |
| LMArena Multi-Turn | 1371 | 1436 |
| EQ-Bench Creative Writing | — | 1611 |
| WildBench | 78.9% | — |
| EQ-Bench 4 | — | 1226 |
Frequently asked questions
Is GLM-4.5-Air better than Inkling?
Inkling is the stronger model overall, scoring 44.1 to 38.9 on the Noometry Index. GLM-4.5-Air costs 6.1× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Which is cheaper, GLM-4.5-Air or Inkling?
GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; Inkling lists at $1.87 and $4.68.
Is GLM-4.5-Air or Inkling better for coding?
Inkling scores higher on coding benchmarks: 34.5 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
GLM-4.5-Air does, with 131K tokens against 66K.
How many benchmarks do GLM-4.5-Air and Inkling share?
17 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and Inkling has 41.