Model comparison

DeepSeek V3 vs GLM-5

Data verified July 17, 2026

Head-to-head evidence from 18 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

DeepSeek V3

DeepSeek

44.84/100

Margin

21.1pts

winning →

GLM-5

Z.AI

65.98/100

1 category wins3 category wins

Public leaderboard positions: DeepSeek V3 #144 (Supported); GLM-5 #26 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. DeepSeek V3 and GLM-5 share 18 comparable benchmark results. 4 of 8 categories are comparable. 5 results are unique to DeepSeek V3; 32 to GLM-5.

Updated July 17, 2026

Shared results: 18
DeepSeek V3 only: 5
GLM-5 only: 32
Comparable categories: 4 / 8

Pick GLM-5 if you want the stronger benchmark profile. DeepSeek V3 only becomes the better choice if knowledge is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 18 shared benchmark results across 7 evidence categories; 4 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

GLM-5 is clearly ahead on the BenchAlign aggregate, 65.98 to 44.84. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GLM-5's sharpest advantage is in mathematics, where it averages 56.3 against 1.7. The single biggest benchmark swing on the page is SWE-bench Verified, 42% to 77.8%. DeepSeek V3 does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.

GLM-5 is also the more expensive model on tokens at $1.00 input / $3.20 output per 1M tokens, versus $0.27 input / $1.10 output per 1M tokens for DeepSeek V3. That is roughly 2.9x on output cost alone. GLM-5 gives you the larger context window at 200K, compared with 128K for DeepSeek V3.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for DeepSeek V3 and GLM-5
Category	DeepSeek V3	Δ	GLM-5
Math	DeepSeek V31.7	Margin→ 54.6	GLM-556.3
Coding	DeepSeek V338.9	Margin→ 27.4	GLM-566.3
Inst. Following	DeepSeek V386.1	Margin→ 6.5	GLM-592.6
Knowledge	DeepSeek V372.8	Margin← 6.2	GLM-566.6
Agentic	DeepSeek V3Not measured	MarginNo overlap	GLM-556.2
Reasoning	DeepSeek V3Not measured	MarginNo overlap	GLM-560.8
Multilingual	DeepSeek V3Not measured	MarginNo overlap	GLM-583.1

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

A · DeepSeek V3B · GLM-5

SWE-bench Verified
Coding
Source ↗
A 42%B 77.8%
Winner: GLM-5Δ 35.8
SWE-bench Verified: DeepSeek V3 scored 42%; GLM-5 scored 77.8%. GLM-5 wins this benchmark.
GPQA
Knowledge
Source ↗
A 59.1%B 86%
Winner: GLM-5Δ 26.9
GPQA: DeepSeek V3 scored 59.1%; GLM-5 scored 86%. GLM-5 wins this benchmark.
FrontierMath v2 (Tiers 1-3)
Math
Source ↗
A 1.724%B 16.434%
Winner: GLM-5Δ 14.7
FrontierMath v2 (Tiers 1-3): DeepSeek V3 scored 1.724%; GLM-5 scored 16.434%. GLM-5 wins this benchmark.
MMLU-Pro
Knowledge
Source ↗
A 75.9%B 85.7%
Winner: GLM-5Δ 9.8
MMLU-Pro: DeepSeek V3 scored 75.9%; GLM-5 scored 85.7%. GLM-5 wins this benchmark.
IFEval
Inst. Following
Source ↗
A 86.1%B 92.6%
Winner: GLM-5Δ 6.5
IFEval: DeepSeek V3 scored 86.1%; GLM-5 scored 92.6%. GLM-5 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

Metric	DeepSeek V3	GLM-5	Comparison
Input / output priceUSD per 1M tokens	DeepSeek V3$0.27 input / $1.1 output	GLM-5$1 input / $3.2 output	DeepSeek V3 has the lower combined listed price.
Generation speedtokens per second	DeepSeek V3Not available	GLM-574 tok/s	A complete speed comparison is not available.
First-answer latencyseconds to first token	DeepSeek V3Not available	GLM-51.64 s	A complete latency comparison is not available.
Context windowmaximum listed tokens	DeepSeek V3128K	GLM-5200K	GLM-5 lists the larger context window.

Benchmark Deep Dive

Agentic

16 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
AA Agentic IndexSource	1.6%	—	Not comparable
τ²-bench resultsSource	22.8%	98.2%	GLM-5 leads
GDPval-AASource	0.0%	—	Not comparable
GDPval-AASource	217	—	Not comparable
Terminal-Bench 2.0Source	—	56.2%	Not comparable
Claw-EvalSource	—	57.7%	Not comparable
QwenClawBenchSource	—	54.1%	Not comparable
τ³-bench resultsSource	—	65.6%	Not comparable
DeepPlanningSource	—	14.6%	Not comparable
ToolathlonSource	—	38%	Not comparable
MCP AtlasSource	—	31.1%	Not comparable
MCP-TasksSource	—	60.8%	Not comparable
WideResearchSource	—	69.8%	Not comparable
CyberGymSource	—	43.2%	Not comparable
APEX-Agents-AASource	—	14.5%	Not comparable
Gert LabsSource	—	50.99%	Not comparable

CodingGLM-5 wins

10 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
LiveCodeBenchSource	37.6%	—	Not comparable
SWE-bench VerifiedSource	42%	77.8%	GLM-5 leads
AA Coding IndexSource	23.0%	—	Not comparable
Terminal-Bench HardSource	6.8%	43.2%	GLM-5 leads
AA-SciCodeSource	35.4%	46.2%	GLM-5 leads
SWE-bench Verified*Source	—	72.8%	Not comparable
SWE-bench ProSource	—	55.1%	Not comparable
SWE MultilingualSource	—	73.3%	Not comparable
SWE-RebenchSource	—	62.8%	Not comparable
React Native EvalsSource	—	74.8%	Not comparable

Reasoning

4 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
AA-LCRSource	29.0%	63.3%	GLM-5 leads
CritPtSource	0.0%	2.0%	GLM-5 leads
LongBench v2Source	—	60.8%	Not comparable
AI-NeedleSource	—	63.3%	Not comparable

KnowledgeDeepSeek V3 wins

12 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
GPQASource	59.1%	86%	GLM-5 leads
MMLU-ProSource	75.9%	85.7%	GLM-5 leads
Artificial Analysis Intelligence IndexSource	14.2%	39.5%	GLM-5 leads
AA-GPQA DiamondSource	55.7%	82.0%	GLM-5 leads
AA-HLESource	3.6%	27.2%	GLM-5 leads
AA-Omniscience IndexSource	-41.3%	2.0%	GLM-5 leads
AA-Omniscience AccuracySource	25.4%	26.9%	GLM-5 leads
AA-Omniscience Hallucination RateSource	89.4%	34.0%	GLM-5 leads
GPQA-DSource	—	86.0%	Not comparable
SuperGPQASource	—	66.8%	Not comparable
MMLU-Pro (Arcee)Source	—	85.8%	Not comparable
HLESource	—	50.4%	Not comparable

MathGLM-5 wins

8 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
FrontierMath v2 (Tiers 1-3)Source	1.724%	16.434%	GLM-5 leads
AIME26Source	—	95.8%	Not comparable
AIME25 (Arcee)Source	—	93.3%	Not comparable
HMMT Feb 2025Source	—	97.5%	Not comparable
HMMT Nov 2025Source	—	96.9%	Not comparable
HMMT Feb 2026Source	—	86.4%	Not comparable
MMAnswerBenchSource	—	82.5%	Not comparable
FrontierMath v2 (Tier 4)Source	—	2.100%	Not comparable

Multilingual

2 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
MMLU-ProXSource	—	83.1%	Not comparable
NOVA-63Source	—	55.1%	Not comparable

Multimodal

1 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
Design Arena WebsiteSource	1154	1282	GLM-5 leads

Inst. FollowingGLM-5 wins

2 benchmarks

Benchmark	DeepSeek V3	GLM-5	Result
IFEvalSource	86.1%	92.6%	GLM-5 leads
AA-IFBenchSource	34.8%	72.3%	GLM-5 leads

Frequently Asked Questions (5)

Which is better, DeepSeek V3 or GLM-5?

GLM-5 is ahead on BenchLM's BenchAlign leaderboard, 65.98 to 44.84. The biggest single separator in this matchup is SWE-bench Verified, where the scores are 42% and 77.8%.

Which is better for knowledge tasks, DeepSeek V3 or GLM-5?

DeepSeek V3 has the edge for knowledge tasks in this comparison, averaging 72.8 versus 66.6. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.

Which is better for coding, DeepSeek V3 or GLM-5?

GLM-5 has the edge for coding in this comparison, averaging 66.3 versus 38.9. Inside this category, Terminal-Bench Hard is the benchmark that creates the most daylight between them.

Which is better for math, DeepSeek V3 or GLM-5?

GLM-5 has the edge for math in this comparison, averaging 56.3 versus 1.7. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.

Which is better for instruction following, DeepSeek V3 or GLM-5?

GLM-5 has the edge for instruction following in this comparison, averaging 92.6 versus 86.1. Inside this category, AA-IFBench is the benchmark that creates the most daylight between them.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

DeepSeek V3

API / mo$1,028

Self-host / mo$18,221

Break-even1.2B/day

GLM-5

API / mo$3,150

Self-host / moNot listed

Break-even—

Proprietary model — self-hosting not applicable.

Model the full break-even

Related Comparisons

Explore More

DeepSeek Compare Pricing Methodology Find Your Best LLM Overall Rankings

Last updated: July 17, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.