Enter opens the highlighted result. Esc closes.

DeepSeek V4.1 Flash vs Qwen3.8 27B

Qwen3.8 27B leads DeepSeek V4.1 Flash on coding, 74.2 to 62.5 on the Vibe Coding Index, and DeepSeek V4.1 Flash costs 3.0x less per output token ($0.84 vs $2.55 per million). Updated October 22, 2026; 75 benchmarks have a result for both.

Replace DeepSeek V4.1 Flash Replace Qwen3.8 27B
DeepSeek V4.1 FlashDeepSeek
Coding
62.5
Overall
62.4
Output $/M
$0.84
67
Qwen3.8 27BQwen
Coding
74.2
Overall
59.6
Output $/M
$2.55
7
vs boards won of 75 shared, 1 tied

DeepSeek V4.1 Flash or Qwen3.8 27B: which to pick

Pick Qwen3.8 27B for coding: it scores 74.2 on the Vibe Coding Index against 62.5 for DeepSeek V4.1 Flash, though on 4 benchmarks against 5.

Pick DeepSeek V4.1 Flash for output tokens 3.0x cheaper ($0.84 vs $2.55 per million) and a larger context window (1.05M against 1M tokens).

Model ADeepSeek V4.1 FlashDeepSeek62.5/1005 independent benchmarksNo cost per solved taskModel BQwen3.8 27BQwen74.2/1004 independent benchmarks$1.26 per solved task
Shared results
75
DeepSeek V4.1 Flash only
26
Qwen3.8 27B only
10
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
DeepSeek V4.1 Flash 62.5 from 5 benchmarks, Qwen3.8 27B 74.2 from 4 benchmarks.
Widest category
Math, 15.5 points apart on average over 3 shared benchmarks.
Widest single gap
LiveBench browser: DeepSeek V4.1 Flash 84.0% against Qwen3.8 27B 32.0%, 52.0 points apart on a 0 to 100 scale.
Output price
DeepSeek V4.1 Flash $0.84 against Qwen3.8 27B $2.55 per million output tokens, 3.0x apart.
Context window
DeepSeek V4.1 Flash takes 1.05M tokens, Qwen3.8 27B 1M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryDeepSeek V4.1 FlashBarsQwen3.8 27BTies
Coding150
Intelligence11
Agentic work531 tie
Reasoning90
Math30
Knowledge51
Web design10
General282
All shared boards6771 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index150
Coding benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
LiveBench coding90.2%86.8%DeepSeek V4.1 Flash
LMArena WebDev16181593DeepSeek V4.1 Flash
AA Coding IndexNot measured68.1Only Qwen3.8 27B measured
LiveCodeBench (Vals AI)Not measured84.0%Only Qwen3.8 27B measured
SWE-bench Verified (Vals AI)Not measured86.0%Only Qwen3.8 27B measured
Terminal-Bench 4 (Vals AI)19.7%Not measuredOnly DeepSeek V4.1 Flash measured
Vibe Code Bench (Vals AI)84.7%64.8%DeepSeek V4.1 Flash
Code migration (Vals AI)45.6%14.2%DeepSeek V4.1 Flash
ProgramBench (Vals AI)0.5%0.0%DeepSeek V4.1 Flash
IOI (Vals AI)40.3%39.1%DeepSeek V4.1 Flash
Terminal-Bench 2.1 (Vals AI)74.5%58.4%DeepSeek V4.1 Flash
SWE-bench Multimodal (Vals AI)46.5%Not measuredOnly DeepSeek V4.1 Flash measured
Vibe Code Bench 1-100 (Vals AI)16.4%Not measuredOnly DeepSeek V4.1 Flash measured
SkillsBench (Vals AI)69.8%38.1%DeepSeek V4.1 Flash
CyberBench (Vals AI)73.7%67.8%DeepSeek V4.1 Flash
SciCode51.9%46.6%DeepSeek V4.1 Flash
FrontierCode43.9%Not measuredOnly DeepSeek V4.1 Flash measured
ALE-Bench1092.3Not measuredOnly DeepSeek V4.1 Flash measured
LMArena Text: coding15021483DeepSeek V4.1 Flash
LMArena Text: software and IT14961478DeepSeek V4.1 Flash
LMArena WebDev: React16281602DeepSeek V4.1 Flash
LMArena WebDev: HTML16121570DeepSeek V4.1 Flash
LMArena WebDev: from an image16061577DeepSeek V4.1 Flash
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
Intelligence benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
AA Intelligence Index24.733.7Qwen3.8 27B
Epoch Capabilities Index155.9151.7DeepSeek V4.1 Flash
Agentic workLike-for-likePublic boards, shown as published, kept out of the index53
Agentic work benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
AA Agentic IndexNot measured45.8Only Qwen3.8 27B measured
LiveBench research86.8%59.4%DeepSeek V4.1 Flash
LiveBench browser84.0%32.0%DeepSeek V4.1 Flash
LiveBench long agentic40.1%27.9%DeepSeek V4.1 Flash
SREBench (Vals AI)0.8%Not measuredOnly DeepSeek V4.1 Flash measured
Finance Agent (Vals AI)53.5%48.5%DeepSeek V4.1 Flash
Tax Agent Bench (Vals AI)26.6%30.9%Qwen3.8 27B
Legal Research Bench (Vals AI)41.4%36.1%DeepSeek V4.1 Flash
Excel Modeling Benchmark (Vals AI)57.2%59.7%Qwen3.8 27B
BioMysteryBench (Vals AI)67.8%Not measuredOnly DeepSeek V4.1 Flash measured
Terminal-Bench-Science (Vals AI)0.0%0.0%Tie
Vals Science Index29.6%Not measuredOnly DeepSeek V4.1 Flash measured
APEX-Agents39.5%47.5%Qwen3.8 27B
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index90
Reasoning benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
GPQA DiamondNot measured88.9%Only Qwen3.8 27B measured
LiveBench reasoning86.7%80.0%DeepSeek V4.1 Flash
MysteryMechanism (Vals AI)21.2%Not measuredOnly DeepSeek V4.1 Flash measured
MMMU Pro (Vals AI)Not measured83.9%Only Qwen3.8 27B measured
GPQA Diamond (Artificial Analysis run)Not measured90.5%Only Qwen3.8 27B measured
Humanity's Last Exam (Artificial Analysis run)39.2%33.9%DeepSeek V4.1 Flash
GPQA Diamond (Epoch AI run)89.8%Not measuredOnly DeepSeek V4.1 Flash measured
ARC-AGI94.5%87.5%DeepSeek V4.1 Flash
ARC-AGI-272.9%42.4%DeepSeek V4.1 Flash
Chess puzzles48.0%Not measuredOnly DeepSeek V4.1 Flash measured
Mystery game puzzles43.0%Not measuredOnly DeepSeek V4.1 Flash measured
LMArena Text: hard prompts14831459DeepSeek V4.1 Flash
LMArena Text: expert15011479DeepSeek V4.1 Flash
LMArena Text: hard prompts in English14881461DeepSeek V4.1 Flash
LMArena Vision: diagrams12971285DeepSeek V4.1 Flash
LMArena Vision: homework13071295DeepSeek V4.1 Flash
Neither model has a result on the other 6 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index30
Math benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
OTIS Mock AIME98.3%Not measuredOnly DeepSeek V4.1 Flash measured
FrontierMath, tiers 1 to 367.4%Not measuredOnly DeepSeek V4.1 Flash measured
FrontierMath, tier 426.8%Not measuredOnly DeepSeek V4.1 Flash measured
ProofBench54.0%16.0%DeepSeek V4.1 Flash
LMArena Text: math14781455DeepSeek V4.1 Flash
LMArena Text: mathematical work14851458DeepSeek V4.1 Flash
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index51
Knowledge benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
MMLU ProNot measured84.3%Only Qwen3.8 27B measured
LegalBench (Vals AI)83.3%82.4%DeepSeek V4.1 Flash
TaxEval (Vals AI)Not measured70.8%Only Qwen3.8 27B measured
MortgageTax (Vals AI)Not measured64.9%Only Qwen3.8 27B measured
MedCode (Vals AI)41.2%28.7%DeepSeek V4.1 Flash
SimpleQA Verified42.9%Not measuredOnly DeepSeek V4.1 Flash measured
LMArena Text: science14691467DeepSeek V4.1 Flash
LMArena Text: medicine and healthcare14841437DeepSeek V4.1 Flash
LMArena Text: legal and government14431447Qwen3.8 27B
LMArena Text: business and finance14611438DeepSeek V4.1 Flash
Neither model has a result on the other 8 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index10
Web design benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
Design Arena: websites1308Not measuredOnly DeepSeek V4.1 Flash measured
Design Arena: UI components1312Not measuredOnly DeepSeek V4.1 Flash measured
Design Arena: game dev1330Not measuredOnly DeepSeek V4.1 Flash measured
Design Arena: data viz1291Not measuredOnly DeepSeek V4.1 Flash measured
Design Arena: 3D1355Not measuredOnly DeepSeek V4.1 Flash measured
Design Arena: all code categories1320Not measuredOnly DeepSeek V4.1 Flash measured
Design Arena: SVG1245Not measuredOnly DeepSeek V4.1 Flash measured
Design Arena: ASCII art1265Not measuredOnly DeepSeek V4.1 Flash measured
LiveBench design79.2%75.9%DeepSeek V4.1 Flash
GeneralLike-for-likePublic boards, shown as published, kept out of the index282
General benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
LiveBench data analysis79.3%76.6%DeepSeek V4.1 Flash
Public Benefits Bench (Vals AI)64.3%Not measuredOnly DeepSeek V4.1 Flash measured
SAGE (Vals AI)47.9%52.4%Qwen3.8 27B
MedScribe (Vals AI)85.5%83.8%DeepSeek V4.1 Flash
Long context reasoning (AA-LCR)84.0%82.0%DeepSeek V4.1 Flash
DTBench89.9%88.0%DeepSeek V4.1 Flash
LMCA47.0%41.4%DeepSeek V4.1 Flash
Surface Evolver Bench46.3%45.0%DeepSeek V4.1 Flash
Furniture assembly34.2%Not measuredOnly DeepSeek V4.1 Flash measured
LMArena Text14641440DeepSeek V4.1 Flash
LMArena Text: instruction following14731436DeepSeek V4.1 Flash
LMArena Text: creative writing14371383DeepSeek V4.1 Flash
LMArena Vision12781269DeepSeek V4.1 Flash
LMArena Text: English14731448DeepSeek V4.1 Flash
LMArena Text: non-English14491428DeepSeek V4.1 Flash
LMArena Text: Chinese15011501Qwen3.8 27B
LMArena Text: German14701443DeepSeek V4.1 Flash
LMArena Text: Japanese14321387DeepSeek V4.1 Flash
LMArena Text: Korean14361384DeepSeek V4.1 Flash
LMArena Text: French14651465DeepSeek V4.1 Flash
LMArena Text: Spanish14591449DeepSeek V4.1 Flash
LMArena Text: Russian14671415DeepSeek V4.1 Flash
LMArena Text: Polish14571439DeepSeek V4.1 Flash
LMArena Text: multi-turn14611440DeepSeek V4.1 Flash
LMArena Text: longer queries14731448DeepSeek V4.1 Flash
LMArena Text: writing and language14501405DeepSeek V4.1 Flash
LMArena Text: entertainment and media14451393DeepSeek V4.1 Flash
LMArena Vision: English12881267DeepSeek V4.1 Flash
LMArena Vision: Chinese13271326DeepSeek V4.1 Flash
LMArena Vision: reading text12821281DeepSeek V4.1 Flash
LMArena Vision: humor12601235DeepSeek V4.1 Flash
LMArena Vision: creative writing12751237DeepSeek V4.1 Flash
Neither model has a result on the other 14 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V4.1 Flash and Qwen3.8 27B
BenchmarkDeepSeek V4.1 FlashQwen3.8 27BLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

DeepSeek V4.1 Flash$0.0006Fits in one request
Qwen3.8 27B$0.0017Fits in one request

DeepSeek V4.1 Flash costs less

Repository review

50K input + 3K output tokens

DeepSeek V4.1 Flash$0.014Fits in one request
Qwen3.8 27B$0.029Fits in one request

DeepSeek V4.1 Flash costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4.1 Flash$0.053Fits in one request
Qwen3.8 27B$0.111Fits in one request

DeepSeek V4.1 Flash costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of DeepSeek V4.1 Flash and Qwen3.8 27B
SpecificationDeepSeek V4.1 FlashQwen3.8 27B
MakerDeepSeekQwen
WeightsOpen weightsOpen weights
ReleasedSeptember 10, 2026August 14, 2026
Context window1.05M1M
Input $/Mtok$0.2225$0.425
Output $/Mtok$0.84$2.55
Coding index62.574.2
Independent measurements behind it54
Cost per solved taskNot computable$1.26
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4.1 Flash better than Qwen3.8 27B for coding?

Pick Qwen3.8 27B for coding: it scores 74.2 on the Vibe Coding Index against 62.5 for DeepSeek V4.1 Flash, though on 4 benchmarks against 5. Pick DeepSeek V4.1 Flash for output tokens 3.0x cheaper ($0.84 vs $2.55 per million) and a larger context window (1.05M against 1M tokens).

Which is cheaper, DeepSeek V4.1 Flash or Qwen3.8 27B?

DeepSeek V4.1 Flash costs $0.2225 per million input tokens and $0.84 per million output tokens; Qwen3.8 27B costs $0.425 and $2.55. Cost per solved task is known for Qwen3.8 27B only.

Which has the bigger context window, DeepSeek V4.1 Flash or Qwen3.8 27B?

DeepSeek V4.1 Flash, with 1.05M tokens against 1M.

How many benchmarks is the DeepSeek V4.1 Flash vs Qwen3.8 27B comparison based on?

75 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4.1 Flash's coding index rests on 5 independent benchmarks and Qwen3.8 27B's on 4. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/deepseek-v4-1-flash-vs-qwen3-8-27b" title="DeepSeek V4.1 Flash vs Qwen3.8 27B, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/deepseek-v4-1-flash-vs-qwen3-8-27b">LLMBenchmark.ai, DeepSeek V4.1 Flash vs Qwen3.8 27B</a></p>