LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GLM 5.3 Flash vs Qwen3.5-35B-A3B

Updated October 21, 2026. 39 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GLM 5.3 Flash Replace Qwen3.5-35B-A3B
GLM 5.3 FlashZ.ai
Coding
71.0
Overall
41.8
Output $/M
$0.50
39
Qwen3.5-35B-A3BQwen
Coding
7.7
Overall
19.3
Output $/M
$1.00
0
vs boards won of 39 shared

Decision reading

Pick GLM 5.3 Flash for coding: it scores 71.0 on the Vibe Coding Index against 7.7 for Qwen3.5-35B-A3B.

Qwen3.5-35B-A3B is neither cheaper nor larger in context, and leads no benchmark both models were measured on.

Model AGLM 5.3 FlashZ.ai71.0/1002 independent benchmarks$0.02 per solved taskModel BQwen3.5-35B-A3BQwen7.7/1001 independent benchmarkNo cost per solved task
Shared results
39
GLM 5.3 Flash only
63
Qwen3.5-35B-A3B only
6
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
GLM 5.3 Flash 71.0 from 2 benchmarks, Qwen3.5-35B-A3B 7.7 from 1 benchmark.
Widest category
Coding, 31.6 points apart on average over 6 shared benchmarks.
Widest single gap
LMArena WebDev: React: GLM 5.3 Flash 1624 against Qwen3.5-35B-A3B 1238, 60.3 points apart on a 0 to 100 scale.
Output price
GLM 5.3 Flash $0.50 against Qwen3.5-35B-A3B $1.00 per million output tokens, 2.0x apart.
Context window
GLM 5.3 Flash takes 1.05M tokens, Qwen3.5-35B-A3B 262K.

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
CategoryGLM 5.3 FlashBarsQwen3.5-35B-A3BTies
Coding60
Intelligence20
Reasoning60
Math30
Knowledge40
General180
All shared boards390
CodingLike-for-likeFour of these boards feed the Vibe Coding Index60
Coding benchmarks for GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
LiveBench coding80.6%Not measuredOnly GLM 5.3 Flash measured
LMArena WebDev16071254GLM 5.3 Flash
AA Coding Index71.5Not measuredOnly GLM 5.3 Flash measured
LiveCodeBench (Vals AI)80.5%Not measuredOnly GLM 5.3 Flash measured
SWE-bench Verified (Vals AI)92.0%Not measuredOnly GLM 5.3 Flash measured
Terminal-Bench 4 (Vals AI)37.4%Not measuredOnly GLM 5.3 Flash measured
Vibe Code Bench (Vals AI)30.8%Not measuredOnly GLM 5.3 Flash measured
Code migration (Vals AI)20.5%Not measuredOnly GLM 5.3 Flash measured
ProgramBench (Vals AI)0.0%Not measuredOnly GLM 5.3 Flash measured
IOI (Vals AI)52.5%Not measuredOnly GLM 5.3 Flash measured
Terminal-Bench 2.1 (Vals AI)62.9%Not measuredOnly GLM 5.3 Flash measured
Vibe Code Bench 1-100 (Vals AI)16.0%Not measuredOnly GLM 5.3 Flash measured
SkillsBench (Vals AI)40.2%Not measuredOnly GLM 5.3 Flash measured
CyberBench (Vals AI)67.9%Not measuredOnly GLM 5.3 Flash measured
SciCode51.6%Not measuredOnly GLM 5.3 Flash measured
Terminal-Bench Hard (Artificial Analysis run)Not measured26.5%Only Qwen3.5-35B-A3B measured
FrontierCode31.8%Not measuredOnly GLM 5.3 Flash measured
FrontierSWE18.1%Not measuredOnly GLM 5.3 Flash measured
ALE-Bench303.6221.8GLM 5.3 Flash
LMArena Text: coding15071410GLM 5.3 Flash
LMArena Text: software and IT15001415GLM 5.3 Flash
LMArena WebDev: React16241238GLM 5.3 Flash
LMArena WebDev: HTML15771301GLM 5.3 Flash
LMArena WebDev: from an image1578Not measuredOnly GLM 5.3 Flash measured
Neither model has a result on the other 14 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
AA Intelligence Index41.819.3GLM 5.3 Flash
Epoch Capabilities Index153.1142.5GLM 5.3 Flash
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index14 boards, none shared
Agentic work benchmarks for GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
AA Agentic Index50.9Not measuredOnly GLM 5.3 Flash measured
LiveBench research70.1%Not measuredOnly GLM 5.3 Flash measured
LiveBench browser84.0%Not measuredOnly GLM 5.3 Flash measured
LiveBench long agentic28.4%Not measuredOnly GLM 5.3 Flash measured
SREBench (Vals AI)0.8%Not measuredOnly GLM 5.3 Flash measured
Finance Agent (Vals AI)57.9%Not measuredOnly GLM 5.3 Flash measured
Tax Agent Bench (Vals AI)31.6%Not measuredOnly GLM 5.3 Flash measured
Legal Research Bench (Vals AI)45.2%Not measuredOnly GLM 5.3 Flash measured
Harvey's Legal Agent Benchmark (Vals AI)10.4%Not measuredOnly GLM 5.3 Flash measured
Excel Modeling Benchmark (Vals AI)55.9%Not measuredOnly GLM 5.3 Flash measured
Terminal-Bench-Science (Vals AI)2.9%Not measuredOnly GLM 5.3 Flash measured
τ²-Bench TelecomNot measured89.2%Only Qwen3.5-35B-A3B measured
APEX-Agents52.8%Not measuredOnly GLM 5.3 Flash measured
Vending-Bench 2Not measured$463Only Qwen3.5-35B-A3B measured
Neither model has a result on the other 25 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
GPQA Diamond86.4%Not measuredOnly GLM 5.3 Flash measured
LiveBench reasoning77.6%Not measuredOnly GLM 5.3 Flash measured
MMMU Pro (Vals AI)86.0%Not measuredOnly GLM 5.3 Flash measured
GPQA Diamond (Artificial Analysis run)91.2%84.5%GLM 5.3 Flash
Humanity's Last Exam (Artificial Analysis run)39.9%21.0%GLM 5.3 Flash
GPQA Diamond (Epoch AI run)90.2%83.5%GLM 5.3 Flash
ARC-AGI91.0%Not measuredOnly GLM 5.3 Flash measured
ARC-AGI-265.8%Not measuredOnly GLM 5.3 Flash measured
Chess puzzles21.0%Not measuredOnly GLM 5.3 Flash measured
Mystery game puzzles8.0%Not measuredOnly GLM 5.3 Flash measured
LMArena Text: hard prompts14901400GLM 5.3 Flash
LMArena Text: expert15111408GLM 5.3 Flash
LMArena Text: hard prompts in English14881411GLM 5.3 Flash
LMArena Vision: diagrams1317Not measuredOnly GLM 5.3 Flash measured
LMArena Vision: homework1318Not measuredOnly GLM 5.3 Flash measured
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 GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
OTIS Mock AIME93.9%70.0%GLM 5.3 Flash
FrontierMath, tiers 1 to 355.8%Not measuredOnly GLM 5.3 Flash measured
FrontierMath, tier 417.1%Not measuredOnly GLM 5.3 Flash measured
ProofBench21.0%Not measuredOnly GLM 5.3 Flash measured
LMArena Text: math15031404GLM 5.3 Flash
LMArena Text: mathematical work14971410GLM 5.3 Flash
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
Knowledge benchmarks for GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
MMLU Pro86.1%Not measuredOnly GLM 5.3 Flash measured
LegalBench (Vals AI)83.9%Not measuredOnly GLM 5.3 Flash measured
TaxEval (Vals AI)75.6%Not measuredOnly GLM 5.3 Flash measured
LMArena Text: science14841413GLM 5.3 Flash
LMArena Text: medicine and healthcare14751421GLM 5.3 Flash
LMArena Text: legal and government14731399GLM 5.3 Flash
LMArena Text: business and finance14601394GLM 5.3 Flash
Neither model has a result on the other 10 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
Web design benchmarks for GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
Design Arena: websites1272Not measuredOnly GLM 5.3 Flash measured
Design Arena: UI components1316Not measuredOnly GLM 5.3 Flash measured
Design Arena: game dev1279Not measuredOnly GLM 5.3 Flash measured
Design Arena: data viz1262Not measuredOnly GLM 5.3 Flash measured
Design Arena: 3D1316Not measuredOnly GLM 5.3 Flash measured
Design Arena: all code categories1282Not measuredOnly GLM 5.3 Flash measured
Design Arena: SVG1284Not measuredOnly GLM 5.3 Flash measured
Design Arena: ASCII art1269Not measuredOnly GLM 5.3 Flash measured
LiveBench design77.2%Not measuredOnly GLM 5.3 Flash measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index180
General benchmarks for GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
Vals Index50.9%Not measuredOnly GLM 5.3 Flash measured
LiveBench data analysis76.4%Not measuredOnly GLM 5.3 Flash measured
MedScribe (Vals AI)88.9%Not measuredOnly GLM 5.3 Flash measured
IFBenchNot measured72.5%Only Qwen3.5-35B-A3B measured
Long context reasoning (AA-LCR)80.0%72.0%GLM 5.3 Flash
DTBenchNot measured80.0%Only Qwen3.5-35B-A3B measured
LMCANot measured29.5%Only Qwen3.5-35B-A3B measured
Surface Evolver Bench52.5%Not measuredOnly GLM 5.3 Flash measured
LMArena Text14711395GLM 5.3 Flash
LMArena Text: instruction following14781379GLM 5.3 Flash
LMArena Text: creative writing14411347GLM 5.3 Flash
LMArena Vision1294Not measuredOnly GLM 5.3 Flash measured
LMArena Text: English14741407GLM 5.3 Flash
LMArena Text: non-English14621378GLM 5.3 Flash
LMArena Text: Chinese15311457GLM 5.3 Flash
LMArena Text: German14621367GLM 5.3 Flash
LMArena Text: Japanese14441326GLM 5.3 Flash
LMArena Text: Korean14361355GLM 5.3 Flash
LMArena Text: French14911412GLM 5.3 Flash
LMArena Text: Spanish14751390GLM 5.3 Flash
LMArena Text: Russian14651376GLM 5.3 Flash
LMArena Text: Polish14501385GLM 5.3 Flash
LMArena Text: multi-turn14691390GLM 5.3 Flash
LMArena Text: longer queries14811389GLM 5.3 Flash
LMArena Text: writing and language14551365GLM 5.3 Flash
LMArena Text: entertainment and media14401347GLM 5.3 Flash
LMArena Vision: English1295Not measuredOnly GLM 5.3 Flash measured
LMArena Vision: Chinese1363Not measuredOnly GLM 5.3 Flash measured
LMArena Vision: reading text1299Not measuredOnly GLM 5.3 Flash measured
LMArena Vision: humor1279Not measuredOnly GLM 5.3 Flash measured
LMArena Vision: creative writing1270Not measuredOnly GLM 5.3 Flash measured
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 GLM 5.3 Flash and Qwen3.5-35B-A3B
BenchmarkGLM 5.3 FlashQwen3.5-35B-A3BLeader
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

GLM 5.3 Flash$0.0004Fits in one request
Qwen3.5-35B-A3B$0.0006Fits in one request

GLM 5.3 Flash costs less

Repository review

50K input + 3K output tokens

GLM 5.3 Flash$0.0090Fits in one request
Qwen3.5-35B-A3B$0.010Fits in one request

GLM 5.3 Flash costs less

Agent loop

200K input + 10K output tokens

GLM 5.3 Flash$0.035Fits in one request
Qwen3.5-35B-A3B$0.038Fits in one request

GLM 5.3 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 GLM 5.3 Flash and Qwen3.5-35B-A3B
SpecificationZ.ai GLM 5.3 FlashQwen3.5-35B-A3B
MakerZ.aiQwen
WeightsOpen weightsOpen weights
ReleasedAugust 26, 2026February 25, 2026
Context window1.05M262K
Input $/Mtok$0.15$0.14
Output $/Mtok$0.50$1.00
Coding index71.07.7
Independent measurements behind it21
Cost per solved task$0.02Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 5.3 Flash better than Qwen3.5-35B-A3B for coding?

Pick GLM 5.3 Flash for coding: it scores 71.0 on the Vibe Coding Index against 7.7 for Qwen3.5-35B-A3B. Qwen3.5-35B-A3B is neither cheaper nor larger in context, and leads no benchmark both models were measured on.

Which is cheaper, GLM 5.3 Flash or Qwen3.5-35B-A3B?

GLM 5.3 Flash costs $0.15 per million input tokens and $0.50 per million output tokens; Qwen3.5-35B-A3B costs $0.14 and $1.00. Cost per solved task is known for GLM 5.3 Flash only.

Which has the bigger context window, GLM 5.3 Flash or Qwen3.5-35B-A3B?

GLM 5.3 Flash, with 1.05M tokens against 262K.

How many benchmarks is the GLM 5.3 Flash vs Qwen3.5-35B-A3B comparison based on?

39 benchmarks have a result for both models as of October 21, 2026. GLM 5.3 Flash's coding index rests on 2 independent benchmarks and Qwen3.5-35B-A3B's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next