LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GLM 4.7 vs Qwen2.5 Coder 32B Instruct

Updated October 21, 2026. 28 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GLM 4.7 Replace Qwen2.5 Coder 32B Instruct
GLM 4.7Z.ai
Coding
33.6
Overall
22.2
Output $/M
$1.98
28
Qwen2.5 Coder 32B InstructQwen
Coding
30.5
Overall
6.7
Output $/M
$1.00
0
vs boards won of 28 shared

Decision reading

Pick GLM 4.7 for coding: it scores 33.6 on the Vibe Coding Index against 30.5 for Qwen2.5 Coder 32B Instruct, though on 1 benchmark against 2.

Pick Qwen2.5 Coder 32B Instruct when output tokens cost 2.0x less.

Model AGLM 4.7Z.ai33.6/1001 independent benchmark$0.29 per solved taskModel BQwen2.5 Coder 32B InstructQwen30.5/1002 independent benchmarksNo cost per solved task
Shared results
28
GLM 4.7 only
52
Qwen2.5 Coder 32B Instruct only
7
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 4.7 33.6 from 1 benchmark, Qwen2.5 Coder 32B Instruct 30.5 from 2 benchmarks.
Widest category
Coding, 40.6 points apart on average over 3 shared benchmarks.
Widest single gap
LiveCodeBench (Artificial Analysis run): GLM 4.7 89.4% against Qwen2.5 Coder 32B Instruct 29.5%, 59.9 points apart on a 0 to 100 scale.
Output price
GLM 4.7 $1.98 against Qwen2.5 Coder 32B Instruct $1.00 per million output tokens, 2.0x apart.
Context window
GLM 4.7 takes 205K tokens, Qwen2.5 Coder 32B Instruct 33K.

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 4.7BarsQwen2.5 Coder 32B InstructTies
Coding30
Intelligence20
Reasoning50
Math20
Knowledge50
General110
All shared boards280
CodingLike-for-likeFour of these boards feed the Vibe Coding Index30
Coding benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
Aider polyglotNot measured16.4%Only Qwen2.5 Coder 32B Instruct measured
SWE-bench VerifiedNot measured47.0%Only Qwen2.5 Coder 32B Instruct measured
LMArena WebDev1435Not measuredOnly GLM 4.7 measured
AA Coding Index45.3Not measuredOnly GLM 4.7 measured
LiveCodeBench (Vals AI)82.2%Not measuredOnly GLM 4.7 measured
SWE-bench Verified (Vals AI)69.4%Not measuredOnly GLM 4.7 measured
LiveCodeBench (Artificial Analysis run)89.4%29.5%GLM 4.7
Terminal-Bench Hard (Artificial Analysis run)31.8%Not measuredOnly GLM 4.7 measured
Terminal-Bench33.4%Not measuredOnly GLM 4.7 measured
ALE-Bench399.5Not measuredOnly GLM 4.7 measured
LMArena Text: coding14541276GLM 4.7
LMArena Text: software and IT14531261GLM 4.7
LMArena WebDev: React1420Not measuredOnly GLM 4.7 measured
LMArena WebDev: HTML1447Not measuredOnly GLM 4.7 measured
Neither model has a result on the other 24 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
AA Intelligence Index22.26.7GLM 4.7
Epoch Capabilities Index143.6119.5GLM 4.7
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Agentic work benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
Design Arena: agents building full-stack apps1050Not measuredOnly GLM 4.7 measured
Design Arena: agents building mobile apps1105Not measuredOnly GLM 4.7 measured
Design Arena: agents building Android apps1118Not measuredOnly GLM 4.7 measured
Design Arena: agents building Godot games1058Not measuredOnly GLM 4.7 measured
τ²-Bench Telecom95.9%Not measuredOnly GLM 4.7 measured
Vending-Bench 2$2,377Not measuredOnly GLM 4.7 measured
Neither model has a result on the other 33 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
GPQA Diamond80.0%Not measuredOnly GLM 4.7 measured
GPQA Diamond (Artificial Analysis run)85.9%41.7%GLM 4.7
Humanity's Last Exam (Artificial Analysis run)27.4%3.5%GLM 4.7
GPQA Diamond (Epoch AI run)83.3%Not measuredOnly GLM 4.7 measured
SimpleBench47.7%Not measuredOnly GLM 4.7 measured
Chess puzzles21.0%Not measuredOnly GLM 4.7 measured
LMArena Text: hard prompts14431251GLM 4.7
LMArena Text: expert14241221GLM 4.7
LMArena Text: hard prompts in English14521269GLM 4.7
Neither model has a result on the other 12 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
AIME93.3%Not measuredOnly GLM 4.7 measured
MGSM (Vals AI)88.2%Not measuredOnly GLM 4.7 measured
AA Math Index95.0Not measuredOnly GLM 4.7 measured
AIME 2025 (Artificial Analysis run)95.0%Not measuredOnly GLM 4.7 measured
OTIS Mock AIME83.3%Not measuredOnly GLM 4.7 measured
ProofBench6.0%Not measuredOnly GLM 4.7 measured
GSM8KNot measured93.0%Only Qwen2.5 Coder 32B Instruct measured
LMArena Text: math14241251GLM 4.7
LMArena Text: mathematical work14111243GLM 4.7
Neither model has a result on the other 4 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index50
Knowledge benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
MMLU Pro82.7%Not measuredOnly GLM 4.7 measured
LegalBench (Vals AI)83.4%Not measuredOnly GLM 4.7 measured
CaseLaw (Vals AI)54.9%Not measuredOnly GLM 4.7 measured
CorpFin (Vals AI)46.4%Not measuredOnly GLM 4.7 measured
TaxEval (Vals AI)68.8%Not measuredOnly GLM 4.7 measured
MedQA (Vals AI)93.7%Not measuredOnly GLM 4.7 measured
MedCode (Vals AI)32.8%Not measuredOnly GLM 4.7 measured
MMLU Pro (Artificial Analysis run)85.6%63.5%GLM 4.7
SimpleQA Verified32.2%Not measuredOnly GLM 4.7 measured
MMLUNot measured79.1%Only Qwen2.5 Coder 32B Instruct measured
AI2 Reasoning ChallengeNot measured70.5%Only Qwen2.5 Coder 32B Instruct measured
LMArena Text: science14571221GLM 4.7
LMArena Text: medicine and healthcare14691187GLM 4.7
LMArena Text: legal and government14441257GLM 4.7
LMArena Text: business and finance14291224GLM 4.7
Neither model has a result on the other 2 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
Design Arena: websites1226Not measuredOnly GLM 4.7 measured
Design Arena: UI components1199Not measuredOnly GLM 4.7 measured
Design Arena: game dev1189Not measuredOnly GLM 4.7 measured
Design Arena: data viz1195Not measuredOnly GLM 4.7 measured
Design Arena: 3D1191Not measuredOnly GLM 4.7 measured
Design Arena: all code categories1217Not measuredOnly GLM 4.7 measured
Design Arena: SVG1142Not measuredOnly GLM 4.7 measured
Design Arena: ASCII art1165Not measuredOnly GLM 4.7 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index110
General benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
MedScribe (Vals AI)68.6%Not measuredOnly GLM 4.7 measured
IFBench67.9%Not measuredOnly GLM 4.7 measured
Long context reasoning (AA-LCR)71.0%Not measuredOnly GLM 4.7 measured
CL-bench15.9%Not measuredOnly GLM 4.7 measured
HellaSwagNot measured83.0%Only Qwen2.5 Coder 32B Instruct measured
WinoGrandeNot measured80.8%Only Qwen2.5 Coder 32B Instruct measured
CL-bench Life10.9%Not measuredOnly GLM 4.7 measured
LMArena Text14351230GLM 4.7
LMArena Text: instruction following14121223GLM 4.7
LMArena Text: creative writing14011174GLM 4.7
LMArena Text: English14531248GLM 4.7
LMArena Text: non-English14171205GLM 4.7
LMArena Text: Chinese14951222GLM 4.7
LMArena Text: German1426Not measuredOnly GLM 4.7 measured
LMArena Text: Japanese1439Not measuredOnly GLM 4.7 measured
LMArena Text: Korean1399Not measuredOnly GLM 4.7 measured
LMArena Text: French1432Not measuredOnly GLM 4.7 measured
LMArena Text: Spanish1434Not measuredOnly GLM 4.7 measured
LMArena Text: Russian14231228GLM 4.7
LMArena Text: Polish1435Not measuredOnly GLM 4.7 measured
LMArena Text: multi-turn14461222GLM 4.7
LMArena Text: longer queries14321251GLM 4.7
LMArena Text: writing and language14111191GLM 4.7
LMArena Text: entertainment and media14001150GLM 4.7
Neither model has a result on the other 22 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GLM 4.7 and Qwen2.5 Coder 32B Instruct
BenchmarkGLM 4.7Qwen2.5 Coder 32B InstructLeader
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 4.7$0.0015Fits in one request
Qwen2.5 Coder 32B Instruct$0.0012Fits in one request

Qwen2.5 Coder 32B Instruct costs less

Repository review

50K input + 3K output tokens

GLM 4.7$0.033Fits in one request
Qwen2.5 Coder 32B Instruct$0.036Exceeds the context window

GLM 4.7 costs less

Agent loop

200K input + 10K output tokens

GLM 4.7$0.128Exceeds the context window
Qwen2.5 Coder 32B Instruct$0.142Exceeds the context window

GLM 4.7 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 4.7 and Qwen2.5 Coder 32B Instruct
SpecificationZ.ai GLM 4.7Qwen2.5 Coder 32B Instruct
MakerZ.aiQwen
WeightsOpen weightsOpen weights
ReleasedDecember 22, 2025November 11, 2024
Context window205K33K
Input $/Mtok$0.54$0.66
Output $/Mtok$1.98$1.00
Coding index33.630.5
Independent measurements behind it12
Cost per solved task$0.29Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GLM 4.7 better than Qwen2.5 Coder 32B Instruct for coding?

Pick GLM 4.7 for coding: it scores 33.6 on the Vibe Coding Index against 30.5 for Qwen2.5 Coder 32B Instruct, though on 1 benchmark against 2. Pick Qwen2.5 Coder 32B Instruct when output tokens cost 2.0x less.

Which is cheaper, GLM 4.7 or Qwen2.5 Coder 32B Instruct?

GLM 4.7 costs $0.54 per million input tokens and $1.98 per million output tokens; Qwen2.5 Coder 32B Instruct costs $0.66 and $1.00. Cost per solved task is known for GLM 4.7 only.

Which has the bigger context window, GLM 4.7 or Qwen2.5 Coder 32B Instruct?

GLM 4.7, with 205K tokens against 33K.

How many benchmarks is the GLM 4.7 vs Qwen2.5 Coder 32B Instruct comparison based on?

28 benchmarks have a result for both models as of October 21, 2026. GLM 4.7's coding index rests on 1 independent benchmark and Qwen2.5 Coder 32B Instruct's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next