Model comparison

DeepSeek LLM 67B vs Qwen3.8 Max

Qwen3.8 Max is the stronger model overall, scoring 56.8 to 24.9 on the Noometry Index.

Last verified October 10, 2026. 14 shared benchmarks.

DeepSeek LLM 67B

DeepSeek

24.9

Rank #347 Confirmed

Qwen3.8 Max

Alibaba (Qwen)

56.8

Rank #22 Confirmed

Summary

They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
The widest gap is in math, where Qwen3.8 Max leads 73.2 to 8.7.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 100% for Qwen3.8 Max.
DeepSeek LLM 67B has downloadable open weights; the other is API-only.

Side by side

DeepSeek LLM 67B and Qwen3.8 Max specifications
	DeepSeek LLM 67B	Qwen3.8 Max
Provider	DeepSeek	Alibaba (Qwen)
Noometry Index	24.9	56.8
Released	2023-11-29	2026-08-02
Weights	Open	Proprietary
Context window	—	1M
Max output	—	131K
Input $ / M tokens	—	$2
Output $ / M tokens	—	$6
Results tracked	15	39

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

Category by category

Coding Qwen3.8 Max leads

DeepSeek LLM 67B: 31.9 (#278), Qwen3.8 Max: 53.5 (#29)

Coding benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
LMArena Coding	1096	1502
DeepSWE	—	57.5%
LMArena WebDev	—	1674
FrontierSWE	—	17.8%
SciCode	—	53.2%

Agentic & Tool Use Not comparable

DeepSeek LLM 67B: —, Qwen3.8 Max: 45.4 (#14)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
APEX-Agents	—	63.3%
τ²-bench Banking	—	55.1%
GDP.pdf	—	23.2%

Reasoning Qwen3.8 Max leads

DeepSeek LLM 67B: 16.5 (#304), Qwen3.8 Max: 54.4 (#26)

Reasoning benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
Chess Puzzles	0%	40%
LMArena Hard Prompts	1070	1496
Epoch Capabilities Index	110.5	156.41
NYT Connections (extended)	—	88.3%
CritPt	—	20%
Mystery Game Puzzles	—	38%
DTBench	—	92%
LMCA	—	46.2%

Math Qwen3.8 Max leads

DeepSeek LLM 67B: 8.7 (#324), Qwen3.8 Max: 73.2 (#20)

Math benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
OTIS Mock AIME 2024-2025	0.8%	100%
LMArena Math	1108	1499
FrontierMath (Tiers 1-3)	—	74.7%
FrontierMath Tier 4	—	46.3%
ProofBench	—	58%
MATH Level 5	6.4%	—

Knowledge Qwen3.8 Max leads

DeepSeek LLM 67B: 7.0 (#313), Qwen3.8 Max: 61.7 (#27)

Knowledge benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
GPQA Diamond	24.6%	92.7%
SimpleQA Verified	—	47.3%
LMArena Expert	—	1507

Multimodal Not comparable

DeepSeek LLM 67B: —, Qwen3.8 Max: 37.2 (#75)

Multimodal benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
LMArena Vision	—	1314
Furniture Assembly	—	20%

Multilingual Qwen3.8 Max leads

DeepSeek LLM 67B: 29.4 (#267), Qwen3.8 Max: 56.7 (#18)

Multilingual benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
LMArena Non-English	1073	1472
LMArena Chinese	1132	1538
LMArena French	—	1503
LMArena German	—	1483
LMArena Japanese	—	1467
LMArena Korean	—	1461
LMArena Russian	—	1481
LMArena Spanish	—	1492

Instruction Following Qwen3.8 Max leads

DeepSeek LLM 67B: 55.4 (#277), Qwen3.8 Max: 77.6 (#17)

Instruction Following benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
LMArena Instruction Following	1079	1479

Long Context Qwen3.8 Max leads

DeepSeek LLM 67B: 33.1 (#265), Qwen3.8 Max: 45.6 (#31)

Long Context benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
LMArena Longer Query	1092	1489

Writing & Preference Qwen3.8 Max leads

DeepSeek LLM 67B: 31.6 (#282), Qwen3.8 Max: 67.1 (#30)

Writing & Preference benchmarks
Benchmark	DeepSeek LLM 67B	Qwen3.8 Max
LMArena Text	1105	1483
LMArena Creative Writing	1067	1479
LMArena Multi-Turn	1082	1489

Frequently asked questions

Is DeepSeek LLM 67B better than Qwen3.8 Max?