# DeepSeek V4 Flash vs GLM-4.5

> DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 41.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-flash-vs-glm-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 8 categories and GLM-4.5 in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 27.7.
- The biggest single-benchmark swing is WeirdML: 63% for DeepSeek V4 Flash and 40.6% for GLM-4.5.
- DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.60 / $2.20 for GLM-4.5.
- DeepSeek V4 Flash accepts more context: 1M tokens versus 131K.

## Snapshot

| | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 53.6 | 41.4 |
| Rank | 32 | 125 |
| Context | 1M | 131K |
| Input $/M | $0.15 | $0.60 |
| Output $/M | $0.60 | $2.20 |
| Weights | Open | Open |

## Coding

- DeepSeek V4 Flash: 47.6 (#56)
- GLM-4.5: 41.1 (#121)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| WeirdML | 63% | 40.6% |
| LMArena Coding | 1457 | 1434 |
| ALE-Bench | 1,306 | 344.82 |
| FrontierCode | 18.8% | — |
| SWE-bench Verified (bash only) | — | 54.2% |
| LMArena WebDev | 1581 | — |
| SciCode | 49.9% | — |
| AlgoTune | — | 1.52 |

## Reasoning

- DeepSeek V4 Flash: 53.7 (#25)
- GLM-4.5: 27.7 (#99)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| Kagi LLM Benchmark | 52.2% | 57.9% |
| LMArena Hard Prompts | 1443 | 1430 |
| ARC-AGI-2 | 61.4% | — |
| SimpleBench | 61.1% | — |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | 89% | — |
| CritPt | 16.6% | — |
| Chess Puzzles | 33% | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 90.9% | — |
| LMCA | 41.7% | — |
| Epoch Capabilities Index | 154.51 | — |

## Math

- DeepSeek V4 Flash: 60.0 (#31)
- GLM-4.5: 36.3 (#119)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| LMArena Math | 1427 | 1427 |
| FrontierMath (Tiers 1-3) | 57.5% | — |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| OTIS Mock AIME 2024-2025 | 94.4% | — |
| ProofBench | 56% | — |

## Knowledge

- DeepSeek V4 Flash: 55.5 (#43)
- GLM-4.5: 35.5 (#172)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| LMArena Expert | 1441 | 1433 |
| GPQA Diamond | 91% | — |
| Humanity's Last Exam | — | 8.3% |
| SimpleQA Verified | 33.6% | — |
| Confabulations | — | 11.3% |

## Multilingual

- DeepSeek V4 Flash: 53.2 (#71)
- GLM-4.5: 53.0 (#74)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| LMArena Non-English | 1420 | 1417 |
| LMArena Chinese | 1467 | 1465 |
| LMArena French | 1438 | 1423 |
| LMArena German | 1418 | 1406 |
| LMArena Japanese | 1403 | 1416 |
| LMArena Korean | 1383 | 1382 |
| LMArena Russian | 1428 | 1414 |
| LMArena Spanish | 1437 | 1454 |

## Instruction Following

- DeepSeek V4 Flash: 75.1 (#76)
- GLM-4.5: 74.3 (#100)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| LMArena Instruction Following | 1421 | 1405 |

## Long Context

- DeepSeek V4 Flash: 43.8 (#81)
- GLM-4.5: 38.2 (#196)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| LMArena Longer Query | 1435 | 1413 |
| Fiction.LiveBench | — | 58.3% |

## Writing & Preference

- DeepSeek V4 Flash: 64.0 (#59)
- GLM-4.5: 57.7 (#121)

| Benchmark | DeepSeek V4 Flash | GLM-4.5 |
|---|---|---|
| LMArena Text | 1432 | 1430 |
| LMArena Creative Writing | 1403 | 1395 |
| EQ-Bench Creative Writing | 1559 | 1343 |
| LMArena Multi-Turn | 1449 | 1415 |
| Short-Story Creative Writing | — | 73.4% |

## FAQ

### Is DeepSeek V4 Flash better than GLM-4.5?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 41.4 on the Noometry Index.

### Which is cheaper, DeepSeek V4 Flash or GLM-4.5?

DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GLM-4.5 lists at $0.60 and $2.20.

### Is DeepSeek V4 Flash or GLM-4.5 better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.6 versus 41.1 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4 Flash does, with 1M tokens against 131K.

### How many benchmarks do DeepSeek V4 Flash and GLM-4.5 share?

21 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and GLM-4.5 has 27.