DeepSeek V4 Pro 0423 vs GLM 5.3 Flash

Updated October 21, 2026. 70 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace DeepSeek V4 Pro 0423 Replace GLM 5.3 Flash

DeepSeek V4 Pro 0423DeepSeek

Coding: 35.3
Overall: 30.4
Output $/M: $1.9105

GLM 5.3 FlashZ.ai

Coding: 71.0
Overall: 41.8
Output $/M: $0.50

vs boards won of 70 shared, 1 tied

Decision reading

Pick GLM 5.3 Flash for coding: it scores 71.0 on the Vibe Coding Index against 35.3 for DeepSeek V4 Pro 0423.

Pick DeepSeek V4 Pro 0423 when what matters is LiveCodeBench (Vals AI) or Vibe Code Bench (Vals AI), where it leads.

Model ADeepSeek V4 Pro 0423DeepSeek35.3/1001 independent benchmark$0.57 per solved task Model BGLM 5.3 FlashZ.ai71.0/1002 independent benchmarks$0.02 per solved task

701332

Shared results: 70
DeepSeek V4 Pro 0423 only: 13
GLM 5.3 Flash only: 32
Like-for-like categories: 8 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.

Coding
Vibe Coding Index, four public coding benchmarks
GLM 5.3 Flash
GLM 5.3 Flash scores 71.0 against 35.3.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
GLM 5.3 Flash
GLM 5.3 Flash leads, 41.8 against 30.4.
Same board, both measured
Agentic work
AA Agentic Index
GLM 5.3 Flash
GLM 5.3 Flash leads, 50.9 against 26.3.
Same board, both measured
Reasoning
GPQA Diamond
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 89.4% against 86.4%.
Same board, both measured
Knowledge
MMLU Pro
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 87.3% against 86.1%.
Same board, both measured
Web design
Design Arena: websites
GLM 5.3 Flash
GLM 5.3 Flash leads, 1272 against 1235.
Same board, both measured
General
Vals Index
No call
Only GLM 5.3 Flash has a Vals Index result, 50.9%.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GLM 5.3 Flash
DeepSeek V4 Pro 0423 $0.0019, GLM 5.3 Flash $0.0004 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GLM 5.3 Flash
DeepSeek V4 Pro 0423 $0.57, GLM 5.3 Flash $0.02.
Published attempt cost and pass rate

At a glance

Coding index: DeepSeek V4 Pro 0423 35.3 from 1 benchmark, GLM 5.3 Flash 71.0 from 2 benchmarks.
Widest category: Coding, 14.4 points apart on average over 16 shared benchmarks.
Widest single gap: LMArena WebDev: React: DeepSeek V4 Pro 0423 1441 against GLM 5.3 Flash 1624, 26.6 points apart on a 0 to 100 scale.
Output price: DeepSeek V4 Pro 0423 $1.9105 against GLM 5.3 Flash $0.50 per million output tokens, 3.8x apart.
Context window: Both take 1.05M tokens.

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
Category	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Ties
Coding	5	10	1 tie
Intelligence	0	2
Agentic work	0	5
Reasoning	3	4
Math	1	5
Knowledge	1	6
Web design	0	8
General	3	16
All shared boards	13	56	1 tie

CodingLike-for-likeFour of these boards feed the Vibe Coding Index510

Coding benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
LiveBench coding	Not measured	80.6%	Only GLM 5.3 Flash measured
LMArena WebDev	1447	1607	GLM 5.3 Flash
AA Coding Index	59.4	71.5	GLM 5.3 Flash
LiveCodeBench (Vals AI)	87.5%	80.5%	DeepSeek V4 Pro 0423
SWE-bench Verified (Vals AI)	77.4%	92.0%	GLM 5.3 Flash
Terminal-Bench 4 (Vals AI)	11.1%	37.4%	GLM 5.3 Flash
Vibe Code Bench (Vals AI)	49.9%	30.8%	DeepSeek V4 Pro 0423
Code migration (Vals AI)	26.2%	20.5%	DeepSeek V4 Pro 0423
ProgramBench (Vals AI)	0.0%	0.0%	Tie
IOI (Vals AI)	Not measured	52.5%	Only GLM 5.3 Flash measured
Terminal-Bench 2.1 (Vals AI)	50.2%	62.9%	GLM 5.3 Flash
Vibe Code Bench 1-100 (Vals AI)	Not measured	16.0%	Only GLM 5.3 Flash measured
SkillsBench (Vals AI)	51.3%	40.2%	DeepSeek V4 Pro 0423
CyberBench (Vals AI)	Not measured	67.9%	Only GLM 5.3 Flash measured
SciCode	Not measured	51.6%	Only GLM 5.3 Flash measured
FrontierCode	17.6%	31.8%	GLM 5.3 Flash
FrontierSWE	Not measured	18.1%	Only GLM 5.3 Flash measured
WeirdML	48.9%	Not measured	Only DeepSeek V4 Pro 0423 measured
ALE-Bench	1006.1	303.6	DeepSeek V4 Pro 0423
LMArena Text: coding	1470	1507	GLM 5.3 Flash
LMArena Text: software and IT	1469	1500	GLM 5.3 Flash
LMArena WebDev: React	1441	1624	GLM 5.3 Flash
LMArena WebDev: HTML	1435	1577	GLM 5.3 Flash
LMArena WebDev: from an image	Not measured	1578	Only 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 index02

Intelligence benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
AA Intelligence Index	30.4	41.8	GLM 5.3 Flash
Epoch Capabilities Index	149.1	153.1	GLM 5.3 Flash

Agentic workLike-for-likePublic boards, shown as published, kept out of the index05

Agentic work benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
AA Agentic Index	26.3	50.9	GLM 5.3 Flash
Design Arena: agents building web apps	1000	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: agents building full-stack apps	948	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: agents building Godot games	1059	Not measured	Only DeepSeek V4 Pro 0423 measured
LiveBench research	Not measured	70.1%	Only GLM 5.3 Flash measured
LiveBench browser	Not measured	84.0%	Only GLM 5.3 Flash measured
LiveBench long agentic	Not measured	28.4%	Only GLM 5.3 Flash measured
SREBench (Vals AI)	Not measured	0.8%	Only GLM 5.3 Flash measured
Finance Agent (Vals AI)	44.1%	57.9%	GLM 5.3 Flash
Tax Agent Bench (Vals AI)	25.2%	31.6%	GLM 5.3 Flash
Legal Research Bench (Vals AI)	23.1%	45.2%	GLM 5.3 Flash
Harvey's Legal Agent Benchmark (Vals AI)	Not measured	10.4%	Only GLM 5.3 Flash measured
Excel Modeling Benchmark (Vals AI)	51.6%	55.9%	GLM 5.3 Flash
Terminal-Bench-Science (Vals AI)	Not measured	2.9%	Only GLM 5.3 Flash measured
APEX-Agents	Not measured	52.8%	Only GLM 5.3 Flash measured
Vending-Bench 2	$3,285	Not measured	Only DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 23 boards here

ReasoningLike-for-likePublic boards, shown as published, kept out of the index34

Reasoning benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
GPQA Diamond	89.4%	86.4%	DeepSeek V4 Pro 0423
LiveBench reasoning	Not measured	77.6%	Only GLM 5.3 Flash measured
MMMU Pro (Vals AI)	Not measured	86.0%	Only GLM 5.3 Flash measured
GPQA Diamond (Artificial Analysis run)	Not measured	91.2%	Only GLM 5.3 Flash measured
Humanity's Last Exam (Artificial Analysis run)	Not measured	39.9%	Only GLM 5.3 Flash measured
GPQA Diamond (Epoch AI run)	90.9%	90.2%	DeepSeek V4 Pro 0423
ARC-AGI	Not measured	91.0%	Only GLM 5.3 Flash measured
ARC-AGI-2	Not measured	65.8%	Only GLM 5.3 Flash measured
Chess puzzles	19.0%	21.0%	GLM 5.3 Flash
Mystery game puzzles	17.0%	8.0%	DeepSeek V4 Pro 0423
LMArena Text: hard prompts	1461	1490	GLM 5.3 Flash
LMArena Text: expert	1463	1511	GLM 5.3 Flash
LMArena Text: hard prompts in English	1466	1488	GLM 5.3 Flash
LMArena Vision: diagrams	Not measured	1317	Only GLM 5.3 Flash measured
LMArena Vision: homework	Not measured	1318	Only 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 index15

Math benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
OTIS Mock AIME	96.7%	93.9%	DeepSeek V4 Pro 0423
FrontierMath, tiers 1 to 3	45.3%	55.8%	GLM 5.3 Flash
FrontierMath, tier 4	2.4%	17.1%	GLM 5.3 Flash
ProofBench	16.0%	21.0%	GLM 5.3 Flash
LMArena Text: math	1443	1503	GLM 5.3 Flash
LMArena Text: mathematical work	1455	1497	GLM 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 index16

Knowledge benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
MMLU Pro	87.3%	86.1%	DeepSeek V4 Pro 0423
LegalBench (Vals AI)	80.3%	83.9%	GLM 5.3 Flash
CaseLaw (Vals AI)	59.4%	Not measured	Only DeepSeek V4 Pro 0423 measured
CorpFin (Vals AI)	61.4%	Not measured	Only DeepSeek V4 Pro 0423 measured
TaxEval (Vals AI)	72.1%	75.6%	GLM 5.3 Flash
MedCode (Vals AI)	40.5%	Not measured	Only DeepSeek V4 Pro 0423 measured
SimpleQA Verified	47.0%	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: science	1470	1484	GLM 5.3 Flash
LMArena Text: medicine and healthcare	1466	1475	GLM 5.3 Flash
LMArena Text: legal and government	1464	1473	GLM 5.3 Flash
LMArena Text: business and finance	1443	1460	GLM 5.3 Flash
Neither model has a result on the other 6 boards here

Web designLike-for-likePublic boards, shown as published, kept out of the index08

Web design benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
Design Arena: websites	1235	1272	GLM 5.3 Flash
Design Arena: UI components	1221	1316	GLM 5.3 Flash
Design Arena: game dev	1229	1279	GLM 5.3 Flash
Design Arena: data viz	1203	1262	GLM 5.3 Flash
Design Arena: 3D	1254	1316	GLM 5.3 Flash
Design Arena: all code categories	1240	1282	GLM 5.3 Flash
Design Arena: SVG	1139	1284	GLM 5.3 Flash
Design Arena: ASCII art	1148	1269	GLM 5.3 Flash
LiveBench design	Not measured	77.2%	Only GLM 5.3 Flash measured

GeneralLike-for-likePublic boards, shown as published, kept out of the index316

General benchmarks for DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
Vals Index	Not measured	50.9%	Only GLM 5.3 Flash measured
LiveBench data analysis	Not measured	76.4%	Only GLM 5.3 Flash measured
Public Benefits Bench (Vals AI)	62.9%	Not measured	Only DeepSeek V4 Pro 0423 measured
MedScribe (Vals AI)	75.1%	88.9%	GLM 5.3 Flash
Long context reasoning (AA-LCR)	Not measured	80.0%	Only GLM 5.3 Flash measured
DTBench	90.7%	Not measured	Only DeepSeek V4 Pro 0423 measured
LMCA	41.2%	Not measured	Only DeepSeek V4 Pro 0423 measured
Surface Evolver Bench	40.0%	52.5%	GLM 5.3 Flash
CL-bench Life	13.5%	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text	1451	1471	GLM 5.3 Flash
LMArena Text: instruction following	1445	1478	GLM 5.3 Flash
LMArena Text: creative writing	1446	1441	DeepSeek V4 Pro 0423
LMArena Vision	Not measured	1294	Only GLM 5.3 Flash measured
LMArena Text: English	1459	1474	GLM 5.3 Flash
LMArena Text: non-English	1439	1462	GLM 5.3 Flash
LMArena Text: Chinese	1481	1531	GLM 5.3 Flash
LMArena Text: German	1458	1462	GLM 5.3 Flash
LMArena Text: Japanese	1446	1444	DeepSeek V4 Pro 0423
LMArena Text: Korean	1409	1436	GLM 5.3 Flash
LMArena Text: French	1473	1491	GLM 5.3 Flash
LMArena Text: Spanish	1457	1475	GLM 5.3 Flash
LMArena Text: Russian	1453	1465	GLM 5.3 Flash
LMArena Text: Polish	1461	1450	DeepSeek V4 Pro 0423
LMArena Text: multi-turn	1468	1469	GLM 5.3 Flash
LMArena Text: longer queries	1458	1481	GLM 5.3 Flash
LMArena Text: writing and language	1447	1455	GLM 5.3 Flash
LMArena Text: entertainment and media	1431	1440	GLM 5.3 Flash
LMArena Vision: English	Not measured	1295	Only GLM 5.3 Flash measured
LMArena Vision: Chinese	Not measured	1363	Only GLM 5.3 Flash measured
LMArena Vision: reading text	Not measured	1299	Only GLM 5.3 Flash measured
LMArena Vision: humor	Not measured	1279	Only GLM 5.3 Flash measured
LMArena Vision: creative writing	Not measured	1270	Only GLM 5.3 Flash measured
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 Pro 0423 and GLM 5.3 Flash
Benchmark	DeepSeek V4 Pro 0423	GLM 5.3 Flash	Leader
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 Pro 0423$0.0019Fits in one request

GLM 5.3 Flash$0.0004Fits in one request

GLM 5.3 Flash costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0423$0.053Fits in one request

GLM 5.3 Flash$0.0090Fits in one request

GLM 5.3 Flash costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0423$0.210Fits in one request

GLM 5.3 Flash$0.035Fits 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 DeepSeek V4 Pro 0423 and GLM 5.3 Flash
Specification	DeepSeek V4 Pro 0423	Z.ai GLM 5.3 Flash
Maker	DeepSeek	Z.ai
Weights	Open weights	Open weights
Released	April 24, 2026	August 26, 2026
Context window	1.05M	1.05M
Input $/Mtok	$0.9553	$0.15
Output $/Mtok	$1.9105	$0.50
Coding index	35.3	71.0
Independent measurements behind it	1	2
Cost per solved task	$0.57	$0.02
Source	OpenRouter listing	OpenRouter listing

Questions

Is DeepSeek V4 Pro 0423 better than GLM 5.3 Flash for coding?

Pick GLM 5.3 Flash for coding: it scores 71.0 on the Vibe Coding Index against 35.3 for DeepSeek V4 Pro 0423. Pick DeepSeek V4 Pro 0423 when what matters is LiveCodeBench (Vals AI) or Vibe Code Bench (Vals AI), where it leads.

Which is cheaper, DeepSeek V4 Pro 0423 or GLM 5.3 Flash?

DeepSeek V4 Pro 0423 costs $0.9553 per million input tokens and $1.9105 per million output tokens; GLM 5.3 Flash costs $0.15 and $0.50. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0423 costs $0.57 and GLM 5.3 Flash $0.02.

Which has the bigger context window, DeepSeek V4 Pro 0423 or GLM 5.3 Flash?

Neither: both take 1.05M tokens.

How many benchmarks is the DeepSeek V4 Pro 0423 vs GLM 5.3 Flash comparison based on?

70 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Pro 0423's coding index rests on 1 independent benchmark and GLM 5.3 Flash's on 2. Where only one model has a result, the page says so instead of naming a winner.

Decision reading

Which one for your work

Coding

Intelligence

Agentic work

Reasoning

Knowledge

Web design

General

Long documents

Everyday cost

Cost per solved task

At a glance

Benchmark by benchmark

What each workload costs

Chat turn

Repository review

Agent loop

Specification differences

Questions

Compare next