Enter opens the highlighted result. Esc closes.

Qwen3.5-122B-A10B vs Qwen3.8 27B

Updated October 21, 2026. 47 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Qwen3.5-122B-A10B Replace Qwen3.8 27B
Qwen3.5-122B-A10BQwen
Coding
22.9
Overall
17.7
Output $/M
$2.08
1
Qwen3.8 27BQwen
Coding
73.6
Overall
33.7
Output $/M
$1.35
46
vs boards won of 47 shared

Decision reading

Pick Qwen3.8 27B for coding: it scores 73.6 on the Vibe Coding Index against 22.9 for Qwen3.5-122B-A10B.

Pick Qwen3.5-122B-A10B when the benchmark that matters is LMArena Vision: humor, where it leads.

Model AQwen3.5-122B-A10BQwen22.9/1001 independent benchmarkNo cost per solved taskModel BQwen3.8 27BQwen73.6/1002 independent benchmarks$1.26 per solved task
Shared results
47
Qwen3.5-122B-A10B only
4
Qwen3.8 27B only
38
Like-for-like categories
6 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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
Qwen3.5-122B-A10B 22.9 from 1 benchmark, Qwen3.8 27B 73.6 from 2 benchmarks.
Widest category
Coding, 22.5 points apart on average over 7 shared benchmarks.
Widest single gap
LMArena WebDev: React: Qwen3.5-122B-A10B 1353 against Qwen3.8 27B 1602, 41.2 points apart on a 0 to 100 scale.
Output price
Qwen3.5-122B-A10B $2.08 against Qwen3.8 27B $1.35 per million output tokens, 1.5x apart.
Context window
Qwen3.5-122B-A10B takes 262K 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
CategoryQwen3.5-122B-A10BBarsQwen3.8 27BTies
Coding07
Intelligence01
Reasoning07
Math02
Knowledge04
General125
All shared boards146
CodingLike-for-likeFour of these boards feed the Vibe Coding Index07
Coding benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
LiveBench codingNot measured86.8%Only Qwen3.8 27B measured
LMArena WebDev13611593Qwen3.8 27B
AA Coding Index43.368.1Qwen3.8 27B
LiveCodeBench (Vals AI)Not measured84.0%Only Qwen3.8 27B measured
SWE-bench Verified (Vals AI)Not measured86.0%Only Qwen3.8 27B measured
Vibe Code Bench (Vals AI)Not measured64.8%Only Qwen3.8 27B measured
Code migration (Vals AI)Not measured14.2%Only Qwen3.8 27B measured
ProgramBench (Vals AI)Not measured0.0%Only Qwen3.8 27B measured
IOI (Vals AI)Not measured39.1%Only Qwen3.8 27B measured
Terminal-Bench 2.1 (Vals AI)Not measured58.4%Only Qwen3.8 27B measured
SkillsBench (Vals AI)Not measured38.1%Only Qwen3.8 27B measured
CyberBench (Vals AI)Not measured67.8%Only Qwen3.8 27B measured
SciCode39.7%46.6%Qwen3.8 27B
Terminal-Bench Hard (Artificial Analysis run)31.1%Not measuredOnly Qwen3.5-122B-A10B measured
LMArena Text: coding14361483Qwen3.8 27B
LMArena Text: software and IT14361478Qwen3.8 27B
LMArena WebDev: React13531602Qwen3.8 27B
LMArena WebDev: HTML13511570Qwen3.8 27B
LMArena WebDev: from an imageNot measured1577Only Qwen3.8 27B measured
Neither model has a result on the other 19 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
AA Intelligence Index17.733.7Qwen3.8 27B
Epoch Capabilities IndexNot measured151.7Only Qwen3.8 27B measured
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index11 boards, none shared
Agentic work benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
AA Agentic IndexNot measured45.8Only Qwen3.8 27B measured
LiveBench researchNot measured59.4%Only Qwen3.8 27B measured
LiveBench browserNot measured32.0%Only Qwen3.8 27B measured
LiveBench long agenticNot measured27.9%Only Qwen3.8 27B measured
Finance Agent (Vals AI)Not measured48.5%Only Qwen3.8 27B measured
Tax Agent Bench (Vals AI)Not measured30.9%Only Qwen3.8 27B measured
Legal Research Bench (Vals AI)Not measured36.1%Only Qwen3.8 27B measured
Excel Modeling Benchmark (Vals AI)Not measured59.7%Only Qwen3.8 27B measured
Terminal-Bench-Science (Vals AI)Not measured0.0%Only Qwen3.8 27B measured
τ²-Bench Telecom93.6%Not measuredOnly Qwen3.5-122B-A10B measured
APEX-AgentsNot measured47.5%Only Qwen3.8 27B measured
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index07
Reasoning benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
GPQA DiamondNot measured88.9%Only Qwen3.8 27B measured
LiveBench reasoningNot measured80.0%Only Qwen3.8 27B measured
MMMU Pro (Vals AI)Not measured83.9%Only Qwen3.8 27B measured
GPQA Diamond (Artificial Analysis run)85.7%90.5%Qwen3.8 27B
Humanity's Last Exam (Artificial Analysis run)25.2%33.9%Qwen3.8 27B
ARC-AGINot measured87.5%Only Qwen3.8 27B measured
ARC-AGI-2Not measured42.4%Only Qwen3.8 27B measured
Mystery game puzzles17.0%Not measuredOnly Qwen3.5-122B-A10B measured
LMArena Text: hard prompts14211459Qwen3.8 27B
LMArena Text: expert14311479Qwen3.8 27B
LMArena Text: hard prompts in English14311461Qwen3.8 27B
LMArena Vision: diagrams12491285Qwen3.8 27B
LMArena Vision: homework12611295Qwen3.8 27B
Neither model has a result on the other 8 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
ProofBenchNot measured16.0%Only Qwen3.8 27B measured
LMArena Text: math14311455Qwen3.8 27B
LMArena Text: mathematical work14311458Qwen3.8 27B
Neither model has a result on the other 10 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
Knowledge benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
MMLU ProNot measured84.3%Only Qwen3.8 27B measured
LegalBench (Vals AI)Not measured82.4%Only Qwen3.8 27B measured
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)Not measured28.7%Only Qwen3.8 27B measured
LMArena Text: science14321467Qwen3.8 27B
LMArena Text: medicine and healthcare14231437Qwen3.8 27B
LMArena Text: legal and government14221447Qwen3.8 27B
LMArena Text: business and finance14141438Qwen3.8 27B
Neither model has a result on the other 8 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
Web design benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
LiveBench designNot measured75.9%Only Qwen3.8 27B measured
Neither model has a result on the other 8 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index125
General benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 27BLeader
LiveBench data analysisNot measured76.6%Only Qwen3.8 27B measured
SAGE (Vals AI)Not measured52.4%Only Qwen3.8 27B measured
MedScribe (Vals AI)Not measured83.8%Only Qwen3.8 27B measured
IFBench75.7%Not measuredOnly Qwen3.5-122B-A10B measured
Long context reasoning (AA-LCR)76.3%82.0%Qwen3.8 27B
DTBench84.3%88.0%Qwen3.8 27B
LMCA32.2%41.4%Qwen3.8 27B
Surface Evolver BenchNot measured45.0%Only Qwen3.8 27B measured
LMArena Text14171440Qwen3.8 27B
LMArena Text: instruction following13991436Qwen3.8 27B
LMArena Text: creative writing13681383Qwen3.8 27B
LMArena Vision12441269Qwen3.8 27B
LMArena Text: English14281448Qwen3.8 27B
LMArena Text: non-English14011428Qwen3.8 27B
LMArena Text: Chinese14631501Qwen3.8 27B
LMArena Text: German14251443Qwen3.8 27B
LMArena Text: Japanese13671387Qwen3.8 27B
LMArena Text: Korean13531384Qwen3.8 27B
LMArena Text: French14431465Qwen3.8 27B
LMArena Text: Spanish14251449Qwen3.8 27B
LMArena Text: Russian14001415Qwen3.8 27B
LMArena Text: Polish13981439Qwen3.8 27B
LMArena Text: multi-turn14161440Qwen3.8 27B
LMArena Text: longer queries14101448Qwen3.8 27B
LMArena Text: writing and language13871405Qwen3.8 27B
LMArena Text: entertainment and media13661393Qwen3.8 27B
LMArena Vision: English12491267Qwen3.8 27B
LMArena Vision: Chinese13011326Qwen3.8 27B
LMArena Vision: reading text12511281Qwen3.8 27B
LMArena Vision: humor12431235Qwen3.5-122B-A10B
LMArena Vision: creative writing12201237Qwen3.8 27B
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Qwen3.5-122B-A10B and Qwen3.8 27B
BenchmarkQwen3.5-122B-A10BQwen3.8 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

Qwen3.5-122B-A10B$0.0013Fits in one request
Qwen3.8 27B$0.0007Fits in one request

Qwen3.8 27B costs less

Repository review

50K input + 3K output tokens

Qwen3.5-122B-A10B$0.019Fits in one request
Qwen3.8 27B$0.0060Fits in one request

Qwen3.8 27B costs less

Agent loop

200K input + 10K output tokens

Qwen3.5-122B-A10B$0.073Fits in one request
Qwen3.8 27B$0.021Fits in one request

Qwen3.8 27B 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 Qwen3.5-122B-A10B and Qwen3.8 27B
SpecificationQwen3.5-122B-A10BQwen3.8 27B
MakerQwenQwen
WeightsOpen weightsOpen weights
ReleasedFebruary 25, 2026August 14, 2026
Context window262K1M
Input $/Mtok$0.26$0.04
Output $/Mtok$2.08$1.35
Coding index22.973.6
Independent measurements behind it12
Cost per solved taskNot computable$1.26
SourceOpenRouter listingOpenRouter listing

Questions

Is Qwen3.5-122B-A10B better than Qwen3.8 27B for coding?

Pick Qwen3.8 27B for coding: it scores 73.6 on the Vibe Coding Index against 22.9 for Qwen3.5-122B-A10B. Pick Qwen3.5-122B-A10B when the benchmark that matters is LMArena Vision: humor, where it leads.

Which is cheaper, Qwen3.5-122B-A10B or Qwen3.8 27B?

Qwen3.5-122B-A10B costs $0.26 per million input tokens and $2.08 per million output tokens; Qwen3.8 27B costs $0.04 and $1.35. Cost per solved task is known for Qwen3.8 27B only.

Which has the bigger context window, Qwen3.5-122B-A10B or Qwen3.8 27B?

Qwen3.8 27B, with 1M tokens against 262K.

How many benchmarks is the Qwen3.5-122B-A10B vs Qwen3.8 27B comparison based on?

47 benchmarks have a result for both models as of October 21, 2026. Qwen3.5-122B-A10B's coding index rests on 1 independent benchmark and Qwen3.8 27B's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next