# Qwen3.8 27B vs Qwen3 Max

> Qwen3.8 27B is the stronger model overall, scoring 46.0 to 43.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen3-8-27b-vs-qwen3-max
- Last updated: 2026-10-10
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. Qwen3.8 27B scores higher in 6 categories and Qwen3 Max in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 22.6.
- The biggest single-benchmark swing is NYT Connections (extended): 54.5% for Qwen3.8 27B and 30.1% for Qwen3 Max.
- Qwen3.8 27B is cheaper at $0.99 / $1.49 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- Qwen3.8 27B has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 46.0 | 43.7 |
| Rank | 68 | 87 |
| Context | 262K | 262K |
| Input $/M | $0.99 | $1.20 |
| Output $/M | $1.49 | $6 |
| Weights | Open | Proprietary |

## Coding

- Qwen3.8 27B: 50.5 (#44)
- Qwen3 Max: 43.0 (#93)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1482 | 1456 |
| LMArena WebDev | 1593 | — |
| SciCode | 46.6% | — |
| ALE-Bench | — | 370.45 |

## Agentic & Tool Use

- Qwen3.8 27B: 32.9 (#57)
- Qwen3 Max: —

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| APEX-Agents | 47.5% | — |
| Vending-Bench 2 | — | 71.56 |

## Reasoning

- Qwen3.8 27B: 41.0 (#54)
- Qwen3 Max: 22.6 (#190)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| NYT Connections (extended) | 54.5% | 30.1% |
| LMArena Hard Prompts | 1460 | 1448 |
| DTBench | 88% | 82.1% |
| LMCA | 41.4% | 28.3% |
| Epoch Capabilities Index | 149.38 | 142.38 |
| ARC-AGI-2 | 42.4% | — |
| Kagi LLM Benchmark | — | 72.5% |
| ARC-AGI-1 | 87.5% | — |
| CritPt | 5.4% | — |
| Chess Puzzles | — | 4% |
| Mystery Game Puzzles | — | 5% |
| Surface Evolver Bench | 45% | — |

## Math

- Qwen3.8 27B: 37.1 (#161)
- Qwen3 Max: 38.7 (#131)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Math | 1456 | 1446 |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| ProofBench | 16% | — |
| MATH Level 5 | — | 97.1% |

## Knowledge

- Qwen3.8 27B: 41.6 (#109)
- Qwen3 Max: 48.1 (#78)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Expert | 1482 | 1455 |
| GPQA Diamond | — | 72.6% |
| SimpleQA Verified | — | 48.7% |

## Multimodal

- Qwen3.8 27B: 41.3 (#37)
- Qwen3 Max: —

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Vision | 1271 | — |

## Multilingual

- Qwen3.8 27B: 53.7 (#60)
- Qwen3 Max: 53.7 (#62)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1430 | 1429 |
| LMArena Chinese | 1504 | 1478 |
| LMArena French | 1465 | 1449 |
| LMArena German | 1438 | 1463 |
| LMArena Japanese | 1384 | 1397 |
| LMArena Korean | 1393 | 1399 |
| LMArena Russian | 1415 | 1428 |
| LMArena Spanish | 1448 | 1462 |

## Instruction Following

- Qwen3.8 27B: 75.8 (#53)
- Qwen3 Max: 74.8 (#87)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1439 | 1419 |

## Long Context

- Qwen3.8 27B: 44.3 (#70)
- Qwen3 Max: 41.6 (#134)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1450 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |

## Writing & Preference

- Qwen3.8 27B: 65.8 (#43)
- Qwen3 Max: 62.4 (#76)

| Benchmark | Qwen3.8 27B | Qwen3 Max |
|---|---|---|
| LMArena Text | 1441 | 1439 |
| LMArena Creative Writing | 1384 | 1402 |
| LMArena Multi-Turn | 1441 | 1446 |
| EQ-Bench Creative Writing | 1671 | — |

## FAQ

### Is Qwen3.8 27B better than Qwen3 Max?

Qwen3.8 27B is the stronger model overall, scoring 46.0 to 43.7 on the Noometry Index.

### Which is cheaper, Qwen3.8 27B or Qwen3 Max?

Qwen3.8 27B is cheaper. It lists at $0.99 per million input tokens and $1.49 per million output tokens; Qwen3 Max lists at $1.20 and $6.

### Is Qwen3.8 27B or Qwen3 Max better for coding?

Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 43.0 in the Noometry coding category.

### Which has the bigger context window?

Both accept 262K tokens.

### How many benchmarks do Qwen3.8 27B and Qwen3 Max share?

21 benchmarks have published results for both models. Qwen3.8 27B has 31 scored results on Noometry and Qwen3 Max has 33.