Noometry

Alibaba (Qwen), proprietary

Qwen3 Max

Qwen3 Max by Alibaba (Qwen) ranks 87th of 354 ranked models on the Noometry Index as of October 2026, with a score of 43.7. Its strongest category is multilingual, where it ranks 62nd. API pricing starts at $1.20 per million input tokens and $6 per million output tokens, with a 262K-token context window.

Last verified

Noometry rank
#87 of 354
Index score
43.7
Evidence
Confirmed 33 results
Provider
Alibaba (Qwen)
Released
September 23, 2025
Weights
Proprietary
Reasoning
No
Context window
262K
Max output
66K
Input price
$1.20 / M
Output price
$6 / M
Blended price
$2.40 / M
Output speed
48 tokens/s Kagi
Value
#156 of 219
Knowledge cutoff
April 2025
Input
text

Category scores

Each category score combines every public result we have in that category.

  1. Coding 43.0
  2. Reasoning 22.6
  3. Math 38.7
  4. Knowledge 48.1
  5. Multilingual 53.7
  6. Instruction Following 74.8
  7. Long Context 41.6
  8. Writing & Preference 62.4
CategoryScoreRankResults
Coding43.0#931
Reasoning22.6#1907
Math38.7#1314
Knowledge48.1#783
Multilingual53.7#621
Instruction Following74.8#871
Long Context41.6#1343
Writing & Preference62.4#763

Strengths and weaknesses

Categories where Qwen3 Max places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Multilingual53.7+6.3#62 of 297, top 21%
Writing & Preference62.4+8.6#76 of 312, top 25%
Knowledge48.1+10.8#78 of 314, top 25%

Weakest categories

CategoryScorevs medianRank
Reasoning22.6−1.0#190 of 350, top 55%
Long Context41.6+0.7#134 of 296, top 46%
Math38.7+2.1#131 of 327, top 41%

Closest competitors

The models ranked just above and below Qwen3 Max. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
ERNIE 5.1#8343.8——Compare
GLM-5V-Turbo#8443.8$1.90—Compare
MiniMax-M3#8543.8$0.52—Compare
Grok 4.3#8643.8$1.56—Compare
MiMo-V2-Omni#8843.6$0.18—Compare
Kimi K2.5 Instant#8943.6——Compare
Gemma 4 31B IT#9043.5$0.153Compare
Qwen3 235B-A22B#9143.5$1.2285Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
LMArena Coding1439LMArena2026-10-08
LMArena Coding1456#76 of 294, top 26%LMArena2026-10-08
ALE-Bench370.45#92 of 105, top 88%Epoch AI

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
Vending-Bench 271.56#54 of 60, top 90%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
Kagi LLM Benchmark72.5%#20 of 99, top 21%Kagi LLM Benchmark
Kagi LLM Benchmark55.9%Kagi LLM Benchmark
NYT Connections (extended)11.8%Lech Mazur benchmarks
NYT Connections (extended)30.1%#72 of 91, top 80%2026-01-23Lech Mazur benchmarks
Chess Puzzles4%#101 of 129, top 79%Epoch AI2025-12-10
LMArena Hard Prompts1423LMArena2026-10-08
LMArena Hard Prompts1448#66 of 297, top 23%LMArena2026-10-08
Mystery Game Puzzles5%#72 of 74, top 98%Epoch AI2026-08-27
DTBench82.1%#66 of 151, top 44%Epoch AI
LMCA28.3%#81 of 125, top 65%Epoch AI
Epoch Capabilities Index142.38#99 of 213, top 47%Epoch AI2025-09-24

Math

BenchmarkScorePositionSettingSourceDate
FrontierMath (Tiers 1-3)18.9%#71 of 81, top 88%Epoch AI2026-08-30
OTIS Mock AIME 2024-202573.3%#87 of 173, top 51%Epoch AI2025-10-06
LMArena Math1446#66 of 285, top 24%LMArena2026-10-08
LMArena Math1434LMArena2026-10-08
MATH Level 597.1%#6 of 79, top 8%Epoch AI2025-10-09

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond72.6%#96 of 186, top 52%Epoch AI2025-10-06
SimpleQA Verified48.7%#28 of 77, top 37%Epoch AI2026-08-27
LMArena Expert1455#65 of 273, top 24%LMArena2026-10-08
LMArena Expert1397LMArena2026-10-08

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1429#62 of 297, top 21%LMArena2026-10-08
LMArena Non-English1400LMArena2026-10-08
LMArena Chinese1430LMArena2026-10-08
LMArena Chinese1478#65 of 285, top 23%LMArena2026-10-08
LMArena French1449#71 of 223, top 32%LMArena2026-10-08
LMArena German1463#33 of 231, top 15%LMArena2026-10-08
LMArena German1437LMArena2026-10-08
LMArena Japanese1397#61 of 211, top 29%LMArena2026-10-08
LMArena Korean1376LMArena2026-10-08
LMArena Korean1399#56 of 213, top 27%LMArena2026-10-08
LMArena Russian1417LMArena2026-10-08
LMArena Russian1428#73 of 283, top 26%LMArena2026-10-08
LMArena Spanish1462#33 of 226, top 15%LMArena2026-10-08
LMArena Spanish1421LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1419#76 of 298, top 26%LMArena2026-10-08
LMArena Instruction Following1398LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
Fiction.LiveBench66.7%#23 of 47, top 49%Epoch AI
CL-bench14.5%#17 of 19, top 90%Epoch AI
LMArena Longer Query1410LMArena2026-10-08
LMArena Longer Query1438#68 of 291, top 24%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1413LMArena2026-10-08
LMArena Text1439#63 of 297, top 22%LMArena2026-10-08
LMArena Creative Writing1402#74 of 295, top 26%LMArena2026-10-08
LMArena Creative Writing1381LMArena2026-10-08
LMArena Multi-Turn1434LMArena2026-10-08
LMArena Multi-Turn1446#61 of 295, top 21%LMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
alibaba$1.20$6—2026-10-10
deepinfra$1.20$6$0.242026-10-10

All Alibaba (Qwen) API prices → Estimate your cost →

Compare Qwen3 Max

Other Alibaba (Qwen) models

Frequently asked questions

How good is Qwen3 Max?

Qwen3 Max by Alibaba (Qwen) ranks 87th of 354 ranked models on the Noometry Index as of October 2026, with a score of 43.7. Its strongest category is multilingual, where it ranks 62nd. API pricing starts at $1.20 per million input tokens and $6 per million output tokens, with a 262K-token context window.

How much does Qwen3 Max cost?

Qwen3 Max costs $1.20 per million input tokens and $6 per million output tokens on Alibaba (Qwen)'s own API.

What is Qwen3 Max's context window?

Qwen3 Max accepts up to 262K tokens of input and can write up to 66K tokens in one response.

Is Qwen3 Max open source?

No. Qwen3 Max is proprietary and available only through Alibaba (Qwen)'s API and partner platforms.

How fast is Qwen3 Max?

Qwen3 Max generated about 48 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

What are Qwen3 Max's strengths and weaknesses?

Relative to other ranked models, Qwen3 Max places best in multilingual, writing & preference, knowledge and lowest in reasoning, long context, math.

What is Qwen3 Max best at?

Its best category is multilingual, where it ranks 62nd on Noometry.