Skip to main content
Radar

Keep up with the models you depend on. Follow price changes, retirements, and API updates.Follow the models you depend on.

Follow model changes
BenchLM
DeepSeek logo
Model A
DeepSeek V3

DeepSeek

41.55/100

Supported · Public rank #181

90% interval 23.0–60.1

DeepSeek V3 vs GLM-5.2

Updated September 14, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Z.AI logo
Model B
GLM-5.2

Z.AI

68.19/100

Supported · Public rank #28

90% interval 61.7–74.7

Decision reading

GLM-5.2 has the higher public score, 68.19 versus 41.55, and the 90% score intervals do not overlap.

1 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Share or export

Share on XLinkedInSocial cardCSVJSON

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

  • Long documents

    Prompts that approach the documented context limit

    GLM-5.2

    GLM-5.2 has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    DeepSeek V3

    DeepSeek V3 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

Show secondary and unsupported calls

  • Repository review cost

    50K fresh input + 3K output tokens

    DeepSeek V3

    DeepSeek V3 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    DeepSeek V3 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    DeepSeek V3 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Not enough matched evidence

    The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. DeepSeek V3 does not fit this workload in one request. GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

Shared results
1
DeepSeek V3 only
5
GLM-5.2 only
24
Like-for-like categories
0 / 8

4 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Directional only
DeepSeek V3
36.7
Estimated · #131/153
GLM-5.2
58.5
Supported · #29/153
Basis
BenchAlign lane · 0 vs 6 public rows
Reading
Directional only

Coding

Directional only
DeepSeek V3
39.3
Estimated · #119/152
GLM-5.2
61.0
Supported · #19/152
Basis
BenchAlign lane · 2 vs 8 public rows
Reading
Directional only

Knowledge

Directional only
DeepSeek V3
39.3
Estimated · #139/183
GLM-5.2
60.7
Supported · #35/183
Basis
BenchAlign lane · 2 vs 6 public rows
Reading
Directional only

Instruction following

Directional only
DeepSeek V3
39.9
#104/123
GLM-5.2
89.8
#22/123
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Directional only

Reasoning

Not comparable
DeepSeek V3
41.2
Unranked · 2 rankable rows
GLM-5.2
74.8
Unranked · 2 rankable rows
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
DeepSeek V3
26.2
Unranked · 1 rankable row
GLM-5.2
80.7
Unranked · 4 rankable rows
Basis
Provisional lane · 1 vs 2 weighted rows
Reading
Not comparable

Multilingual

Not comparable
DeepSeek V3
Not ranked
GLM-5.2
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Multimodal

Not comparable
DeepSeek V3
Not ranked
GLM-5.2
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

DeepSeek V3
$0.00082
Fits in one request
GLM-5.2
$0.0036
Fits in one request

DeepSeek V3 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

DeepSeek V3
$0.0168
Fits in one request
GLM-5.2
$0.0832
Fits in one request

DeepSeek V3 has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

DeepSeek V3
$0.0304
Does not fit in one request
GLM-5.2
$0.352
Fits in one request
Cached input priced at the published list-input rate

DeepSeek V3 does not fit this workload in one request. GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate.

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Context window

Maximum documented context; output-token limits may be lower.

DeepSeek V3

128K

GLM-5.2

1M

API model ID

DeepSeek V3

Not sourced

GLM-5.2

Not sourced

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

DeepSeek V3

$0.07 per 1M cached input tokens

GLM-5.2

Not published

Documented inputs

DeepSeek V3

Not sourced

GLM-5.2

Not sourced

Documented outputs

DeepSeek V3

Not sourced

GLM-5.2

Not sourced

Provider availability

DeepSeek V3

Not sourced

GLM-5.2

Not sourced

Reasoning profile

DeepSeek V3

Non-Reasoning

GLM-5.2

Reasoning

Weight access

DeepSeek V3

Open Weight

GLM-5.2

Open Weight

License

DeepSeek V3

Open Weight

GLM-5.2

Open Weight

Release date

DeepSeek V3

2024-12-26

GLM-5.2

2026-06-16

If you already use one of these models
Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
GLM-5.2 has the higher public score, 68.19 versus 41.55, and the 90% score intervals do not overlap.
Workload cost
Repository review: $0.0168 vs $0.0832. Cache-heavy agent loop: $0.0304 vs $0.352.
Context tradeoff
GLM-5.2 has the larger documented window (1M).

Run the same representative tasks against both endpoints before changing production traffic.

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-5.2
API / mo$4,350
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Model the full break-even

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence30 rows

Agentic

  • Terminal-Bench 3.0

    DeepSeek V3—
    GLM-5.24.6%
    Source

    Not directly comparable

  • Terminal-Bench 2.0

    DeepSeek V3—
    GLM-5.281%
    Source

    Not directly comparable

  • MCP Atlas

    DeepSeek V3—
    GLM-5.276.8%
    Source

    Not directly comparable

  • Toolathlon

    DeepSeek V3—
    GLM-5.248.2%
    Source

    Not directly comparable

  • ResearchClawBench

    DeepSeek V3—
    GLM-5.220.7%
    Source

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    DeepSeek V3—
    GLM-5.267.8%
    Source

    Not directly comparable

Coding

  • LiveCodeBench

    DeepSeek V337.6%
    Source
    GLM-5.2—

    Not directly comparable

  • SWE-bench Verified

    DeepSeek V342%
    Source
    GLM-5.2—

    Not directly comparable

  • SWE-bench Pro

    DeepSeek V3—
    GLM-5.262.1%
    Source

    Not directly comparable

  • NL2Repo

    DeepSeek V3—
    GLM-5.248.9%
    Source

    Not directly comparable

  • Terminal-Bench 2.0

    DeepSeek V3—
    GLM-5.281.0%
    Source

    Not directly comparable

  • ProgramBench

    DeepSeek V3—
    GLM-5.263.7%
    Source

    Not directly comparable

  • cursorBench32

    DeepSeek V3—
    GLM-5.255.0%
    Source

    Not directly comparable

  • OpenHarmony Bench

    DeepSeek V3—
    GLM-5.258.4%
    Source

    Not directly comparable

  • LiveCodeBench (Vals)

    DeepSeek V3—
    GLM-5.269.5%
    Source

    Not directly comparable

  • SWE-bench (Vals)

    DeepSeek V3—
    GLM-5.282.8%
    Source

    Not directly comparable

Reasoning

  • CritPt

    DeepSeek V3—
    GLM-5.220.9%
    Source

    Not directly comparable

Knowledge

  • GPQA

    DeepSeek V359.1%
    Source
    GLM-5.291.2%
    Source

    GLM-5.2 leads this result

  • MMLU-Pro

    DeepSeek V375.9%
    Source
    GLM-5.2—

    Not directly comparable

  • GPQA-D

    DeepSeek V3—
    GLM-5.291.2%
    Source

    Not directly comparable

  • HLE

    DeepSeek V3—
    GLM-5.254.7%
    Source

    Not directly comparable

  • HLE w/o tools

    DeepSeek V3—
    GLM-5.240.5%
    Source

    Not directly comparable

  • GPQA Diamond (Vals)

    DeepSeek V3—
    GLM-5.285.6%
    Source

    Not directly comparable

  • MMLU-Pro (Vals)

    DeepSeek V3—
    GLM-5.286.7%
    Source

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    DeepSeek V31.724%
    Source
    GLM-5.2—

    Not directly comparable

  • AIME26

    DeepSeek V3—
    GLM-5.299.2%
    Source

    Not directly comparable

  • HMMT Nov 2025

    DeepSeek V3—
    GLM-5.294.4%
    Source

    Not directly comparable

  • HMMT Feb 2026

    DeepSeek V3—
    GLM-5.292.5%
    Source

    Not directly comparable

  • MMAnswerBench

    DeepSeek V3—
    GLM-5.291.0%
    Source

    Not directly comparable

Instruction following

  • IFEval

    DeepSeek V386.1%
    Source
    GLM-5.2—

    Not directly comparable

Frequently asked questions

Which is better, DeepSeek V3 or GLM-5.2?

GLM-5.2 has the higher public score, 68.19 versus 41.55, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.

Which is better for coding, DeepSeek V3 or GLM-5.2?

GLM-5.2 scores higher for coding on the public lane, 61 to 39.3. DeepSeek V3 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, DeepSeek V3 or GLM-5.2?

GLM-5.2 scores higher for agentic tasks on the public lane, 58.5 to 36.7. DeepSeek V3 is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which costs less, DeepSeek V3 or GLM-5.2?

For the stated presets, chat costs $0.00082 on DeepSeek V3 and $0.0036 on GLM-5.2; repository review costs $0.0168 and $0.0832; the cache-heavy agent loop costs $0.0304 and $0.352. DeepSeek V3 does not fit this workload in one request. GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, DeepSeek V3 or GLM-5.2?

GLM-5.2 has the larger documented context window: 1M, compared with 128K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated September 14, 2026

Watch DeepSeek V3 vs GLM-5.2

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.