Skip to main content
BenchLM
Advertise

Model comparison

DeepSeek V4 Flash vs Qwen3.5 397B

Data verified

Head-to-head evidence from 11 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

58.88/100
Margin
1.9pts
← winning
57.01/100
0 category wins4 category wins

Public leaderboard positions: DeepSeek V4 Flash #61 (Estimated); Qwen3.5 397B #71 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. DeepSeek V4 Flash and Qwen3.5 397B share 11 comparable benchmark results. 4 of 8 categories are comparable. 11 results are unique to DeepSeek V4 Flash; 44 to Qwen3.5 397B.

Updated July 23, 2026
Shared results
11
DeepSeek V4 Flash only
11
Qwen3.5 397B only
44
Comparable categories
4 / 8

Pick DeepSeek V4 Flash if you want the stronger benchmark profile. Qwen3.5 397B only becomes the better choice if mathematics is the priority.

Confidence note. This is a partial-evidence comparison with 11 shared benchmark results across 4 evidence categories; 4 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 Flash has the cleaner BenchAlign overall profile here, landing at 58.88 versus 57.01. It is a real lead, but still close enough that category-level strengths matter more than the headline number.

Qwen3.5 397B is also the more expensive model on tokens at $0.60 input / $3.60 output per 1M tokens, versus $0.14 input / $0.28 output per 1M tokens for DeepSeek V4 Flash. That is roughly 12.9x on output cost alone. DeepSeek V4 Flash 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 Flash and Qwen3.5 397B
CategoryDeepSeek V4 FlashΔQwen3.5 397B
MathDeepSeek V4 Flash40.8Margin→ 49.8Qwen3.5 397B90.6
KnowledgeDeepSeek V4 Flash38.8Margin→ 17.8Qwen3.5 397B56.6
AgenticDeepSeek V4 Flash49.1Margin→ 7.4Qwen3.5 397B56.5
CodingDeepSeek V4 Flash64.2Margin→ 2.3Qwen3.5 397B66.5
ReasoningDeepSeek V4 FlashNot measuredMarginNo overlapQwen3.5 397B63.2
MultilingualDeepSeek V4 FlashNot measuredMarginNo overlapQwen3.5 397B84.7
MultimodalDeepSeek V4 FlashNot measuredMarginNo overlapQwen3.5 397B79.6
Inst. FollowingDeepSeek V4 FlashNot measuredMarginNo overlapQwen3.5 397B92.6

Decisive benchmark drivers

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

More
A · DeepSeek V4 FlashB · Qwen3.5 397B

  1. HMMT Feb 2026

    Math
    Source ↗
    A 40.8%B 87.9%
    Winner: Qwen3.5 397BΔ 47.1
    HMMT Feb 2026: DeepSeek V4 Flash scored 40.8%; Qwen3.5 397B scored 87.9%. Qwen3.5 397B wins this benchmark.

  2. HLE

    Knowledge
    Source ↗
    A 8.1%B 28.7%
    Winner: Qwen3.5 397BΔ 20.6
    HLE: DeepSeek V4 Flash scored 8.1%; Qwen3.5 397B scored 28.7%. Qwen3.5 397B wins this benchmark.

  3. GPQA

    Knowledge
    Source ↗
    A 71.2%B 88.4%
    Winner: Qwen3.5 397BΔ 17.2
    GPQA: DeepSeek V4 Flash scored 71.2%; Qwen3.5 397B scored 88.4%. Qwen3.5 397B wins this benchmark.

  4. MMLU-Pro

    Knowledge
    Source ↗
    A 83%B 87.8%
    Winner: Qwen3.5 397BΔ 4.8
    MMLU-Pro: DeepSeek V4 Flash scored 83%; Qwen3.5 397B scored 87.8%. Qwen3.5 397B wins this benchmark.

  5. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 49.1%B 52.5%
    Winner: Qwen3.5 397BΔ 3.4
    Terminal-Bench 2.0: DeepSeek V4 Flash scored 49.1%; Qwen3.5 397B scored 52.5%. Qwen3.5 397B wins this benchmark.

Operational comparison

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

MetricDeepSeek V4 FlashQwen3.5 397BComparison
Input / output priceUSD per 1M tokensDeepSeek V4 Flash$0.14 input / $0.28 outputQwen3.5 397B$0.6 input / $3.6 outputDeepSeek V4 Flash has the lower combined listed price.
Generation speedtokens per secondDeepSeek V4 FlashNot availableQwen3.5 397B96 tok/sA complete speed comparison is not available.
First-answer latencyseconds to first tokenDeepSeek V4 FlashNot availableQwen3.5 397B2.44 sA complete latency comparison is not available.
Context windowmaximum listed tokensDeepSeek V4 Flash1MQwen3.5 397B128KDeepSeek V4 Flash lists the larger context window.

Benchmark Deep Dive

AgenticQwen3.5 397B wins
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
Terminal-Bench 2.0Source 49.1%52.5%Qwen3.5 397B leads
MCP AtlasSource 64%46.1%DeepSeek V4 Flash leads
ToolathlonSource 40.7%36.3%DeepSeek V4 Flash leads
Claw-EvalSource 57.8%56.8%DeepSeek V4 Flash leads
Gert LabsSource 54.35%46.76%DeepSeek V4 Flash leads
BrowseCompSource —62%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
MCP-TasksSource —74.2%Not comparable
WideResearchSource —74.0%Not comparable
τ²-bench resultsSource —95.6%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 —962Not comparable
CodingQwen3.5 397B wins
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
SWE-bench VerifiedSource 73.7%76.2%Qwen3.5 397B leads
SWE-bench ProSource 49.1%50.9%Qwen3.5 397B leads
SWE MultilingualSource 69.7%—Not comparable
Terminal-Bench 2.0Source 49.1%—Not comparable
LiveCodeBench v6Source —83.6%Not comparable
AA-SciCodeSource —42.0%Not comparable
AA Coding IndexSource —48.2%Not comparable
Reasoning
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
MRCR 1MSource 37.5%—Not comparable
CorpusQA 1MSource 15.5%—Not comparable
LongBench v2Source —63.2%Not comparable
AI-NeedleSource —68.7%Not comparable
AA-LCRSource —65.7%Not comparable
CritPtSource —1.7%Not comparable
KnowledgeQwen3.5 397B wins
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
MMLU-ProSource 83%87.8%Qwen3.5 397B leads
SimpleQASource 23.1%—Not comparable
Chinese-SimpleQASource 71.5%—Not comparable
GPQASource 71.2%88.4%Qwen3.5 397B leads
GPQA-DSource 71.2%—Not comparable
HLESource 8.1%28.7%Qwen3.5 397B leads
SuperGPQASource —70.4%Not comparable
MMLU-ReduxSource —94.9%Not comparable
C-EvalSource —93%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
MathQwen3.5 397B wins
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
HMMT Feb 2026Source 40.8%87.9%Qwen3.5 397B leads
IMOAnswerBenchSource 41.9%—Not comparable
ApexSource 1.0%—Not comparable
Apex ShortlistSource 9.3%—Not comparable
AIME26Source —93.3%Not comparable
HMMT Feb 2025Source —94.8%Not comparable
HMMT Nov 2025Source —92.7%Not comparable
MMAnswerBenchSource —80.9%Not comparable
Multilingual
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
MMLU-ProXSource —84.7%Not comparable
NOVA-63Source —59.1%Not comparable
Multimodal
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
Design Arena WebsiteSource 1238—Not comparable
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
BenchmarkDeepSeek V4 FlashQwen3.5 397BResult
IFEvalSource —92.6%Not comparable
AA-IFBenchSource —78.8%Not comparable
Frequently Asked Questions (5)

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

DeepSeek V4 Flash is ahead on BenchLM's BenchAlign leaderboard, 58.88 to 57.01. The biggest single separator in this matchup is HMMT Feb 2026, where the scores are 40.8% and 87.9%.

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

Qwen3.5 397B has the edge for knowledge tasks in this comparison, averaging 56.6 versus 38.8. Inside this category, HLE is the benchmark that creates the most daylight between them.

Which is better for coding, DeepSeek V4 Flash or Qwen3.5 397B?

Qwen3.5 397B has the edge for coding in this comparison, averaging 66.5 versus 64.2. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for math, DeepSeek V4 Flash or Qwen3.5 397B?

Qwen3.5 397B has the edge for math in this comparison, averaging 90.6 versus 40.8. Inside this category, HMMT Feb 2026 is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, DeepSeek V4 Flash or Qwen3.5 397B?

Qwen3.5 397B has the edge for agentic tasks in this comparison, averaging 56.5 versus 49.1. Inside this category, MCP Atlas 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.