Model comparison

DeepSeek V4 Pro Base vs Qwen3.5 397B

Data verified July 23, 2026

Head-to-head evidence from 5 shared benchmark results across 2 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

DeepSeek V4 Pro Base

DeepSeek

N/A

No comparison

Qwen3.5 397B

Alibaba

57.01/100

1 category wins1 category wins

Public leaderboard positions: DeepSeek V4 Pro Base unranked (Not scored); Qwen3.5 397B #71 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. DeepSeek V4 Pro Base and Qwen3.5 397B share 5 comparable benchmark results. 2 of 8 categories are comparable. 19 results are unique to DeepSeek V4 Pro Base; 50 to Qwen3.5 397B.

Updated July 23, 2026

Shared results: 5
DeepSeek V4 Pro Base only: 19
Qwen3.5 397B only: 50
Comparable categories: 2 / 8

Treat this as a split decision. DeepSeek V4 Pro Base makes more sense if knowledge is the priority or you need the larger 1M context window; Qwen3.5 397B is the better fit if reasoning is the priority.

Confidence note. This is a partial-evidence comparison with 5 shared benchmark results across 2 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

DeepSeek V4 Pro Base and Qwen3.5 397B finish on the same BenchAlign overall score, so this is less about a single winner and more about where the edge shows up. The BenchAlign headline says tie; the benchmark table is where the real choice happens.

DeepSeek V4 Pro Base gives you the larger context window at 1M, compared with 128K for Qwen3.5 397B.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for DeepSeek V4 Pro Base and Qwen3.5 397B
Category	DeepSeek V4 Pro Base	Δ	Qwen3.5 397B
Reasoning	DeepSeek V4 Pro Base51.5	Margin→ 11.7	Qwen3.5 397B63.2
Knowledge	DeepSeek V4 Pro Base66.4	Margin← 9.8	Qwen3.5 397B56.6
Agentic	DeepSeek V4 Pro BaseNot measured	MarginNo overlap	Qwen3.5 397B56.5
Coding	DeepSeek V4 Pro BaseNot measured	MarginNo overlap	Qwen3.5 397B66.5
Math	DeepSeek V4 Pro BaseNot measured	MarginNo overlap	Qwen3.5 397B90.6
Multilingual	DeepSeek V4 Pro BaseNot measured	MarginNo overlap	Qwen3.5 397B84.7
Multimodal	DeepSeek V4 Pro BaseNot measured	MarginNo overlap	Qwen3.5 397B79.6
Inst. Following	DeepSeek V4 Pro BaseNot measured	MarginNo overlap	Qwen3.5 397B92.6

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

A · DeepSeek V4 Pro BaseB · Qwen3.5 397B

SuperGPQA
Knowledge
Source ↗
A 53.9%B 70.4%
Winner: Qwen3.5 397BΔ 16.5
SuperGPQA: DeepSeek V4 Pro Base scored 53.9%; Qwen3.5 397B scored 70.4%. Qwen3.5 397B wins this benchmark.
MMLU-Pro
Knowledge
Source ↗
A 73.5%B 87.8%
Winner: Qwen3.5 397BΔ 14.3
MMLU-Pro: DeepSeek V4 Pro Base scored 73.5%; Qwen3.5 397B scored 87.8%. Qwen3.5 397B wins this benchmark.
LongBench v2
Reasoning
Source ↗
A 51.5%B 63.2%
Winner: Qwen3.5 397BΔ 11.7
LongBench v2: DeepSeek V4 Pro Base scored 51.5%; Qwen3.5 397B scored 63.2%. Qwen3.5 397B wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

Metric	DeepSeek V4 Pro Base	Qwen3.5 397B	Comparison
Input / output priceUSD per 1M tokens	DeepSeek V4 Pro BaseNot available	Qwen3.5 397B$0.6 input / $3.6 output	A complete price comparison is not available.
Generation speedtokens per second	DeepSeek V4 Pro BaseNot available	Qwen3.5 397B96 tok/s	A complete speed comparison is not available.
First-answer latencyseconds to first token	DeepSeek V4 Pro BaseNot available	Qwen3.5 397B2.44 s	A complete latency comparison is not available.
Context windowmaximum listed tokens	DeepSeek V4 Pro Base1M	Qwen3.5 397B128K	DeepSeek V4 Pro Base lists the larger context window.

Benchmark Deep Dive

Agentic

18 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
Terminal-Bench 2.0Source	—	52.5%	Not comparable
BrowseCompSource	—	62%	Not comparable
Claw-EvalSource	—	56.8%	Not comparable
QwenClawBenchSource	—	51.8%	Not comparable
τ³-bench resultsSource	—	68.4%	Not comparable
VITA-BenchSource	—	43.7%	Not comparable
DeepPlanningSource	—	37.6%	Not comparable
ToolathlonSource	—	36.3%	Not comparable
MCP AtlasSource	—	46.1%	Not comparable
MCP-TasksSource	—	74.2%	Not comparable
WideResearchSource	—	74.0%	Not comparable
τ²-bench resultsSource	—	95.6%	Not comparable
Gert LabsSource	—	46.76%	Not comparable
ResearchClawBenchSource	—	14.2%	Not comparable
AA Agentic IndexSource	—	19.9%	Not comparable
APEX-Agents-AASource	—	15.3%	Not comparable
GDPval-AASource	—	23.1%	Not comparable
GDPval-AASource	—	962	Not comparable

Coding

7 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
BigCodeBenchSource	59.2%	—	Not comparable
HumanEvalSource	76.8%	—	Not comparable
SWE-bench VerifiedSource	—	76.2%	Not comparable
LiveCodeBench v6Source	—	83.6%	Not comparable
SWE-bench ProSource	—	50.9%	Not comparable
AA-SciCodeSource	—	42.0%	Not comparable
AA Coding IndexSource	—	48.2%	Not comparable

ReasoningQwen3.5 397B wins

9 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
BBHSource	87.5%	—	Not comparable
DROPSource	88.7%	—	Not comparable
HellaSwagSource	88.0%	—	Not comparable
WinoGrandeSource	81.5%	—	Not comparable
CLUEWSCSource	85.2%	—	Not comparable
LongBench v2Source	51.5%	63.2%	Qwen3.5 397B leads
AI-NeedleSource	—	68.7%	Not comparable
AA-LCRSource	—	65.7%	Not comparable
CritPtSource	—	1.7%	Not comparable

KnowledgeDeepSeek V4 Pro Base wins

20 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
AGIEvalSource	83.1%	—	Not comparable
MMLUSource	90.1%	—	Not comparable
MMLU-ReduxSource	90.8%	94.9%	Qwen3.5 397B leads
MMLU-ProSource	73.5%	87.8%	Qwen3.5 397B leads
MMMLUSource	90.3%	—	Not comparable
C-EvalSource	93.1%	93%	DeepSeek V4 Pro Base leads
CMMLUSource	90.8%	—	Not comparable
MultiLoKoSource	51.1%	—	Not comparable
SimpleQASource	55.2%	—	Not comparable
SuperGPQASource	53.9%	70.4%	Qwen3.5 397B leads
FACTS ParametricSource	62.6%	—	Not comparable
TriviaQASource	85.6%	—	Not comparable
GPQASource	—	88.4%	Not comparable
HLESource	—	28.7%	Not comparable
Artificial Analysis Intelligence IndexSource	—	33.7%	Not comparable
AA-GPQA DiamondSource	—	89.3%	Not comparable
AA-HLESource	—	27.3%	Not comparable
AA-Omniscience IndexSource	—	-29.8%	Not comparable
AA-Omniscience AccuracySource	—	31.4%	Not comparable
AA-Omniscience Hallucination RateSource	—	89.1%	Not comparable

Math

8 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
GSM8KSource	92.6%	—	Not comparable
MATHSource	64.5%	—	Not comparable
CMathSource	90.9%	—	Not comparable
AIME26Source	—	93.3%	Not comparable
HMMT Feb 2025Source	—	94.8%	Not comparable
HMMT Nov 2025Source	—	92.7%	Not comparable
HMMT Feb 2026Source	—	87.9%	Not comparable
MMAnswerBenchSource	—	80.9%	Not comparable

Multilingual

3 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
MGSMSource	84.4%	—	Not comparable
MMLU-ProXSource	—	84.7%	Not comparable
NOVA-63Source	—	59.1%	Not comparable

Multimodal

7 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
MMMU-ProSource	—	79%	Not comparable
MathVisionSource	—	88.6%	Not comparable
CharXivSource	—	80.8%	Not comparable
VideoMMMUSource	—	84.7%	Not comparable
ScreenSpot ProSource	—	65.6%	Not comparable
V*Source	—	95.8%	Not comparable
AA-MMMU-ProSource	—	77.3%	Not comparable

Inst. Following

2 benchmarks

Benchmark	DeepSeek V4 Pro Base	Qwen3.5 397B	Result
IFEvalSource	—	92.6%	Not comparable
AA-IFBenchSource	—	78.8%	Not comparable

Frequently Asked Questions (3)

Which is better, DeepSeek V4 Pro Base or Qwen3.5 397B?

DeepSeek V4 Pro Base and Qwen3.5 397B are tied on the BenchAlign overall score, so the right pick depends on which category matters most for your use case.

Which is better for knowledge tasks, DeepSeek V4 Pro Base or Qwen3.5 397B?

DeepSeek V4 Pro Base has the edge for knowledge tasks in this comparison, averaging 66.4 versus 56.6. Inside this category, SuperGPQA is the benchmark that creates the most daylight between them.

Which is better for reasoning, DeepSeek V4 Pro Base or Qwen3.5 397B?

Qwen3.5 397B has the edge for reasoning in this comparison, averaging 63.2 versus 51.5. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

DeepSeek Compare Pricing Methodology Find Your Best LLM Overall Rankings

Last updated: July 23, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.