# GLM-5 vs Inkling

> GLM-5 is the stronger model overall, scoring 46.1 to 44.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-5-vs-inkling
- Last updated: 2026-10-11
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. GLM-5 scores higher in 6 categories and Inkling in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5 leads 46.4 to 31.3.
- The biggest single-benchmark swing is ARC-AGI-1: 44.7% for GLM-5 and 79.5% for Inkling.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- GLM-5 accepts more context: 205K tokens versus 66K.

## Snapshot

| | GLM-5 | Inkling |
|---|---|---|
| Provider | Z.ai (Zhipu) | Thinking Machines Lab |
| Noometry Index | 46.1 | 44.1 |
| Rank | 66 | 80 |
| Context | 205K | 66K |
| Input $/M | $1 | $1.87 |
| Output $/M | $3.20 | $4.68 |
| Weights | Open | Open |

## Coding

- GLM-5: 49.0 (#52)
- Inkling: 34.5 (#234)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| LMArena WebDev | 1434 | 1413 |
| WeirdML | 48.2% | 32.3% |
| LMArena Coding | 1461 | 1464 |
| ALE-Bench | 765.62 | 946 |
| SWE-bench Verified | 72.1% | — |
| FrontierCode | — | 14% |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 69.7% | — |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |

## Agentic & Tool Use

- GLM-5: 31.1 (#71)
- Inkling: 29.6 (#85)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| τ²-bench Banking | 9.8% | 25% |
| Terminal-Bench | 52.4% | — |
| APEX-Agents | — | 33.8% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| Vending-Bench 2 | 4,432 | — |

## Reasoning

- GLM-5: 27.6 (#116)
- Inkling: 40.4 (#56)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| ARC-AGI-2 | 4.9% | 36.5% |
| SimpleBench | 53.2% | 50% |
| ARC-AGI-1 | 44.7% | 79.5% |
| Chess Puzzles | 10% | 21% |
| LMArena Hard Prompts | 1452 | 1451 |
| Epoch Capabilities Index | 145.83 | 148.54 |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 74.8% | — |
| CritPt | — | 5.4% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| ForecastBench | 61 | — |

## Math

- GLM-5: 46.4 (#71)
- Inkling: 31.3 (#225)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 88.9% |
| LMArena Math | 1440 | 1479 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| MathArena Final-Answer Competitions | 65.7% | — |
| ProofBench | — | 0% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GLM-5: 52.3 (#64)
- Inkling: 55.1 (#49)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| GPQA Diamond | 87.8% | 88.3% |
| LMArena Expert | 1454 | 1465 |
| SimpleQA Verified | — | 40.3% |
| Vectara Hallucination Rate | 10.1% | — |

## Multilingual

- GLM-5: 53.7 (#58)
- Inkling: 54.0 (#52)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| LMArena Non-English | 1430 | 1434 |
| LMArena Chinese | 1511 | 1490 |
| LMArena French | 1455 | 1458 |
| LMArena German | 1445 | 1446 |
| LMArena Japanese | 1416 | 1429 |
| LMArena Korean | 1423 | 1404 |
| LMArena Russian | 1436 | 1429 |
| LMArena Spanish | 1454 | 1448 |

## Instruction Following

- GLM-5: 75.2 (#67)
- Inkling: 75.1 (#71)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| LMArena Instruction Following | 1428 | 1426 |

## Long Context

- GLM-5: 44.7 (#60)
- Inkling: 43.8 (#86)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| LMArena Longer Query | 1446 | 1434 |
| CL-bench | 18.7% | — |

## Writing & Preference

- GLM-5: 66.0 (#38)
- Inkling: 65.2 (#51)

| Benchmark | GLM-5 | Inkling |
|---|---|---|
| LMArena Text | 1446 | 1441 |
| LMArena Creative Writing | 1439 | 1387 |
| EQ-Bench Creative Writing | 1601 | 1611 |
| LMArena Multi-Turn | 1456 | 1436 |
| EQ-Bench 4 | — | 1226 |

## FAQ

### Is GLM-5 better than Inkling?

GLM-5 is the stronger model overall, scoring 46.1 to 44.1 on the Noometry Index.

### Which is cheaper, GLM-5 or Inkling?

GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Inkling lists at $1.87 and $4.68.

### Is GLM-5 or Inkling better for coding?

GLM-5 scores higher on coding benchmarks: 49.0 versus 34.5 in the Noometry coding category.

### Which has the bigger context window?

GLM-5 does, with 205K tokens against 66K.

### How many benchmarks do GLM-5 and Inkling share?

29 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Inkling has 41.