Enter opens the highlighted result. Esc closes.

GLM 5.3 Flash vs Hy3

Updated October 21, 2026. 44 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 Hy3
GLM 5.3 FlashZ.ai
Coding
71.0
Overall
41.8
Output $/M
$0.50
43
Hy3Tencent
Coding
43.9
Overall
No data
Output $/M
$0.528
1
vs boards won of 44 shared

Decision reading

Pick GLM 5.3 Flash for coding: it scores 71.0 on the Vibe Coding Index against 43.9 for Hy3.

Pick Hy3 when the benchmark that matters is ALE-Bench, where it leads.

Model AGLM 5.3 FlashZ.ai71.0/1002 independent benchmarks$0.02 per solved taskModel BHy3Tencent43.9/1001 independent benchmarkNo cost per solved task
Shared results
44
GLM 5.3 Flash only
58
Hy3 only
0
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, Hy3 43.9 from 1 benchmark.
Widest category
Coding, 11.2 points apart on average over 7 shared benchmarks.
Widest single gap
ALE-Bench: GLM 5.3 Flash 303.6 against Hy3 838.6, 17.8 points apart on a 0 to 100 scale.
Output price
GLM 5.3 Flash $0.50 against Hy3 $0.528 per million output tokens, 1.1x apart.
Context window
GLM 5.3 Flash takes 1.05M tokens, Hy3 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 FlashBarsHy3Ties
Coding61
Reasoning50
Math20
Knowledge40
Web design80
General180
All shared boards431
CodingLike-for-likeFour of these boards feed the Vibe Coding Index61
Coding benchmarks for GLM 5.3 Flash and Hy3
BenchmarkGLM 5.3 FlashHy3Leader
LiveBench coding80.6%Not measuredOnly GLM 5.3 Flash measured
LMArena WebDev16071507GLM 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%48.6%GLM 5.3 Flash
FrontierCode31.8%Not measuredOnly GLM 5.3 Flash measured
FrontierSWE18.1%Not measuredOnly GLM 5.3 Flash measured
ALE-Bench303.6838.6Hy3
LMArena Text: coding15071466GLM 5.3 Flash
LMArena Text: software and IT15001463GLM 5.3 Flash
LMArena WebDev: React16241511GLM 5.3 Flash
LMArena WebDev: HTML15771487GLM 5.3 Flash
LMArena WebDev: from an image1578Not measuredOnly GLM 5.3 Flash measured
Neither model has a result on the other 15 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Intelligence benchmarks for GLM 5.3 Flash and Hy3
BenchmarkGLM 5.3 FlashHy3Leader
AA Intelligence Index41.8Not measuredOnly GLM 5.3 Flash measured
Epoch Capabilities Index153.1Not measuredOnly GLM 5.3 Flash measured
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
Agentic work benchmarks for GLM 5.3 Flash and Hy3
BenchmarkGLM 5.3 FlashHy3Leader
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
APEX-Agents52.8%Not measuredOnly GLM 5.3 Flash measured
Neither model has a result on the other 27 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for GLM 5.3 Flash and Hy3
BenchmarkGLM 5.3 FlashHy3Leader
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%89.7%GLM 5.3 Flash
Humanity's Last Exam (Artificial Analysis run)39.9%33.5%GLM 5.3 Flash
GPQA Diamond (Epoch AI run)90.2%Not measuredOnly GLM 5.3 Flash measured
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 prompts14901447GLM 5.3 Flash
LMArena Text: expert15111461GLM 5.3 Flash
LMArena Text: hard prompts in English14881455GLM 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 index20
Math benchmarks for GLM 5.3 Flash and Hy3
BenchmarkGLM 5.3 FlashHy3Leader
OTIS Mock AIME93.9%Not measuredOnly GLM 5.3 Flash measured
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: math15031473GLM 5.3 Flash
LMArena Text: mathematical work14971482GLM 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 Hy3
BenchmarkGLM 5.3 FlashHy3Leader
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: science14841457GLM 5.3 Flash
LMArena Text: medicine and healthcare14751452GLM 5.3 Flash
LMArena Text: legal and government14731422GLM 5.3 Flash
LMArena Text: business and finance14601424GLM 5.3 Flash
Neither model has a result on the other 10 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for GLM 5.3 Flash and Hy3
BenchmarkGLM 5.3 FlashHy3Leader
Design Arena: websites12721230GLM 5.3 Flash
Design Arena: UI components13161205GLM 5.3 Flash
Design Arena: game dev12791208GLM 5.3 Flash
Design Arena: data viz12621163GLM 5.3 Flash
Design Arena: 3D13161233GLM 5.3 Flash
Design Arena: all code categories12821227GLM 5.3 Flash
Design Arena: SVG12841199GLM 5.3 Flash
Design Arena: ASCII art12691157GLM 5.3 Flash
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 Hy3
BenchmarkGLM 5.3 FlashHy3Leader
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
Long context reasoning (AA-LCR)80.0%79.0%GLM 5.3 Flash
Surface Evolver Bench52.5%Not measuredOnly GLM 5.3 Flash measured
LMArena Text14711438GLM 5.3 Flash
LMArena Text: instruction following14781426GLM 5.3 Flash
LMArena Text: creative writing14411403GLM 5.3 Flash
LMArena Vision1294Not measuredOnly GLM 5.3 Flash measured
LMArena Text: English14741448GLM 5.3 Flash
LMArena Text: non-English14621426GLM 5.3 Flash
LMArena Text: Chinese15311485GLM 5.3 Flash
LMArena Text: German14621449GLM 5.3 Flash
LMArena Text: Japanese14441397GLM 5.3 Flash
LMArena Text: Korean14361396GLM 5.3 Flash
LMArena Text: French14911456GLM 5.3 Flash
LMArena Text: Spanish14751457GLM 5.3 Flash
LMArena Text: Russian14651432GLM 5.3 Flash
LMArena Text: Polish14501447GLM 5.3 Flash
LMArena Text: multi-turn14691436GLM 5.3 Flash
LMArena Text: longer queries14811442GLM 5.3 Flash
LMArena Text: writing and language14551414GLM 5.3 Flash
LMArena Text: entertainment and media14401400GLM 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 18 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 Hy3
BenchmarkGLM 5.3 FlashHy3Leader
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
Hy3$0.0004Fits in one request

Hy3 costs less

Repository review

50K input + 3K output tokens

GLM 5.3 Flash$0.0090Fits in one request
Hy3$0.0082Fits in one request

Hy3 costs less

Agent loop

200K input + 10K output tokens

GLM 5.3 Flash$0.035Fits in one request
Hy3$0.032Fits in one request

Hy3 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 Hy3
SpecificationZ.ai GLM 5.3 FlashTencent Hy3
MakerZ.aiTencent
WeightsOpen weightsOpen weights
ReleasedAugust 26, 2026July 6, 2026
Context window1.05M262K
Input $/Mtok$0.15$0.132
Output $/Mtok$0.50$0.528
Coding index71.043.9
Independent measurements behind it21
Cost per solved task$0.02Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 5.3 Flash better than Hy3 for coding?

Pick GLM 5.3 Flash for coding: it scores 71.0 on the Vibe Coding Index against 43.9 for Hy3. Pick Hy3 when the benchmark that matters is ALE-Bench, where it leads.

Which is cheaper, GLM 5.3 Flash or Hy3?

GLM 5.3 Flash costs $0.15 per million input tokens and $0.50 per million output tokens; Hy3 costs $0.132 and $0.528. Cost per solved task is known for GLM 5.3 Flash only.

Which has the bigger context window, GLM 5.3 Flash or Hy3?

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

How many benchmarks is the GLM 5.3 Flash vs Hy3 comparison based on?

44 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 Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next