Skip to main content
BenchLM
Advertise

Model comparison

DeepSeek V4 Flash (High) vs Qwen3.5-122B-A10B

Data verified

Head-to-head evidence from 20 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

53.95/100
Margin
6.6pts
winning →
60.56/100
0 category wins3 category wins

Public leaderboard positions: DeepSeek V4 Flash (High) #92 (Estimated); Qwen3.5-122B-A10B #47 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. DeepSeek V4 Flash (High) and Qwen3.5-122B-A10B share 20 comparable benchmark results. 3 of 8 categories are comparable. 18 results are unique to DeepSeek V4 Flash (High); 11 to Qwen3.5-122B-A10B.

Updated July 23, 2026
Shared results
20
DeepSeek V4 Flash (High) only
18
Qwen3.5-122B-A10B only
11
Comparable categories
3 / 8

Pick Qwen3.5-122B-A10B if you want the stronger benchmark profile. DeepSeek V4 Flash (High) only becomes the better choice if you need the larger 1M context window.

Confidence note. This is a partial-evidence comparison with 20 shared benchmark results across 5 evidence categories; 3 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Qwen3.5-122B-A10B is clearly ahead on the BenchAlign aggregate, 60.56 to 53.95. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Qwen3.5-122B-A10B's sharpest advantage is in knowledge, where it averages 83.6 against 52.1. The single biggest benchmark swing on the page is BrowseComp, 53.5% to 63.8%.

DeepSeek V4 Flash (High) is also the more expensive model on tokens at $0.14 input / $0.28 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3.5-122B-A10B. That is roughly Infinityx on output cost alone. DeepSeek V4 Flash (High) gives you the larger context window at 1M, compared with 262K for Qwen3.5-122B-A10B.

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 (High) and Qwen3.5-122B-A10B
CategoryDeepSeek V4 Flash (High)ΔQwen3.5-122B-A10B
KnowledgeDeepSeek V4 Flash (High)52.1Margin→ 31.5Qwen3.5-122B-A10B83.6
CodingDeepSeek V4 Flash (High)68.5Margin→ 3.5Qwen3.5-122B-A10B72.0
AgenticDeepSeek V4 Flash (High)55.3Margin→ 1.1Qwen3.5-122B-A10B56.4
ReasoningDeepSeek V4 Flash (High)Not measuredMarginNo overlapQwen3.5-122B-A10B60.2
MathDeepSeek V4 Flash (High)91.9MarginNo overlapQwen3.5-122B-A10BNot measured
MultilingualDeepSeek V4 Flash (High)Not measuredMarginNo overlapQwen3.5-122B-A10B82.2
MultimodalDeepSeek V4 Flash (High)Not measuredMarginNo overlapQwen3.5-122B-A10B77.2
Inst. FollowingDeepSeek V4 Flash (High)Not measuredMarginNo overlapQwen3.5-122B-A10B93.4

Decisive benchmark drivers

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

More
A · DeepSeek V4 Flash (High)B · Qwen3.5-122B-A10B

  1. BrowseComp

    Agentic
    Source ↗
    A 53.5%B 63.8%
    Winner: Qwen3.5-122B-A10BΔ 10.3
    BrowseComp: DeepSeek V4 Flash (High) scored 53.5%; Qwen3.5-122B-A10B scored 63.8%. Qwen3.5-122B-A10B wins this benchmark.

  2. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 56.6%B 49.4%
    Winner: DeepSeek V4 Flash (High)Δ 7.2
    Terminal-Bench 2.0: DeepSeek V4 Flash (High) scored 56.6%; Qwen3.5-122B-A10B scored 49.4%. DeepSeek V4 Flash (High) wins this benchmark.

  3. SWE-bench Verified

    Coding
    Source ↗
    A 78.6%B 72%
    Winner: DeepSeek V4 Flash (High)Δ 6.6
    SWE-bench Verified: DeepSeek V4 Flash (High) scored 78.6%; Qwen3.5-122B-A10B scored 72%. DeepSeek V4 Flash (High) wins this benchmark.

  4. GPQA

    Knowledge
    Source ↗
    A 87.4%B 86.6%
    Winner: DeepSeek V4 Flash (High)Δ 0.8
    GPQA: DeepSeek V4 Flash (High) scored 87.4%; Qwen3.5-122B-A10B scored 86.6%. DeepSeek V4 Flash (High) wins this benchmark.

  5. MMLU-Pro

    Knowledge
    Source ↗
    A 86.4%B 86.7%
    Winner: Qwen3.5-122B-A10BΔ 0.3
    MMLU-Pro: DeepSeek V4 Flash (High) scored 86.4%; Qwen3.5-122B-A10B scored 86.7%. Qwen3.5-122B-A10B wins this benchmark.

Operational comparison

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

MetricDeepSeek V4 Flash (High)Qwen3.5-122B-A10BComparison
Input / output priceUSD per 1M tokensDeepSeek V4 Flash (High)$0.14 input / $0.28 outputQwen3.5-122B-A10B$0 input / $0 outputQwen3.5-122B-A10B has the lower combined listed price.
Generation speedtokens per secondDeepSeek V4 Flash (High)Not availableQwen3.5-122B-A10BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenDeepSeek V4 Flash (High)Not availableQwen3.5-122B-A10BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensDeepSeek V4 Flash (High)1MQwen3.5-122B-A10B262KDeepSeek V4 Flash (High) lists the larger context window.

Benchmark Deep Dive

AgenticQwen3.5-122B-A10B wins
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
Terminal-Bench 2.0Source 56.6%49.4%DeepSeek V4 Flash (High) leads
BrowseCompSource 53.5%63.8%Qwen3.5-122B-A10B leads
HLE w/ toolsSource 40.3%—Not comparable
MCP AtlasSource 67.4%—Not comparable
ToolathlonSource 43.5%—Not comparable
τ²-bench resultsSource 95.6%93.6%DeepSeek V4 Flash (High) leads
AA Agentic IndexSource 28.2%20.7%DeepSeek V4 Flash (High) leads
GDPval-AASource 32.4%23.9%DeepSeek V4 Flash (High) leads
GDPval-AASource 1147978DeepSeek V4 Flash (High) leads
OSWorld-VerifiedSource —58%Not comparable
CodingQwen3.5-122B-A10B wins
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
CodeforcesSource 2816.0—Not comparable
SWE-bench VerifiedSource 78.6%72%DeepSeek V4 Flash (High) leads
SWE-bench ProSource 52.3%—Not comparable
SWE MultilingualSource 70.2%—Not comparable
Terminal-Bench 2.0Source 56.6%—Not comparable
AA-SciCodeSource 42.0%42.0%Tie
AA Coding IndexSource 52.0%45.7%DeepSeek V4 Flash (High) leads
Reasoning
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
MRCR 1MSource 76.9%—Not comparable
CorpusQA 1MSource 59.3%—Not comparable
AA-LCRSource 62.7%66.7%Qwen3.5-122B-A10B leads
CritPtSource 3.4%0.6%DeepSeek V4 Flash (High) leads
LongBench v2Source —60.2%Not comparable
KnowledgeQwen3.5-122B-A10B wins
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
MMLU-ProSource 86.4%86.7%Qwen3.5-122B-A10B leads
SimpleQASource 28.9%—Not comparable
Chinese-SimpleQASource 73.2%—Not comparable
GPQASource 87.4%86.6%DeepSeek V4 Flash (High) leads
GPQA-DSource 87.4%—Not comparable
HLESource 29.4%—Not comparable
Artificial Analysis Intelligence IndexSource 37.5%32.3%DeepSeek V4 Flash (High) leads
AA-GPQA DiamondSource 86.7%85.7%DeepSeek V4 Flash (High) leads
AA-HLESource 27.8%23.4%DeepSeek V4 Flash (High) leads
AA-Omniscience IndexSource -22.3%-39.6%DeepSeek V4 Flash (High) leads
AA-Omniscience AccuracySource 35.5%24.7%DeepSeek V4 Flash (High) leads
AA-Omniscience Hallucination RateSource 89.7%85.5%Qwen3.5-122B-A10B leads
SuperGPQASource —67.1%Not comparable
Math
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
HMMT Feb 2026Source 91.9%—Not comparable
IMOAnswerBenchSource 85.1%—Not comparable
ApexSource 19.1%—Not comparable
Apex ShortlistSource 72.1%—Not comparable
Multilingual
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
MMLU-ProXSource —82.2%Not comparable
Multimodal
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
Design Arena WebsiteSource 1238—Not comparable
MMMUSource —83.9%Not comparable
MMVUSource —74.7%Not comparable
MathVisionSource —86.2%Not comparable
CharXivSource —77.2%Not comparable
V*Source —93.2%Not comparable
AA-MMMU-ProSource —75.0%Not comparable
Inst. Following
BenchmarkDeepSeek V4 Flash (High)Qwen3.5-122B-A10BResult
AA-IFBenchSource 73.5%75.7%Qwen3.5-122B-A10B leads
IFEvalSource —93.4%Not comparable
Frequently Asked Questions (4)

Which is better, DeepSeek V4 Flash (High) or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B is ahead on BenchLM's BenchAlign leaderboard, 60.56 to 53.95. The biggest single separator in this matchup is BrowseComp, where the scores are 53.5% and 63.8%.

Which is better for knowledge tasks, DeepSeek V4 Flash (High) or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for knowledge tasks in this comparison, averaging 83.6 versus 52.1. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.

Which is better for coding, DeepSeek V4 Flash (High) or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for coding in this comparison, averaging 72 versus 68.5. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, DeepSeek V4 Flash (High) or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for agentic tasks in this comparison, averaging 56.4 versus 55.3. Inside this category, GDPval-AA 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.