Skip to main content
Radar

Every change to the models you run, with its source and its date. Releases, price changes, retirements, API changes, and incidents.Every change to the models you run, with its source.

Follow model changes
BenchLM

Claude Opus 4.5 vs GLM-5

Decision reading

GLM-5 has the higher public score estimate, 61.47 versus 58.42, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

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

Anthropic logo
Model A
Claude Opus 4.5

Anthropic

58.42/100

Supported · Public rank #73

90% interval 46.9–70.0

Z.AI logo
Model B
GLM-5

Z.AI

61.47/100

Supported · Public rank #55

90% interval 50.2–72.8

Updated September 18, 2026. Rank says GLM-5 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

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.

  • Chat turn cost

    1K fresh input + 500 output tokens

    GLM-5

    GLM-5 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

    GLM-5

    GLM-5 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

    Claude Opus 4.5 and GLM-5 are 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

    Claude Opus 4.5 and GLM-5 are scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented

  • 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. Claude Opus 4.5 does not fit this workload in one request. GLM-5 does not fit this workload in one request. Claude Opus 4.5 has no published cached-input rate, so cached tokens use its listed input rate. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

56.7Claude Opus 4.556.0GLM-5

Directional only · BenchAlign

Claude Opus 4.5 scores higher, but at least one score rests on Estimated evidence or a different benchmark set. Directional only, no winner.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Open the coding leaderboardShortlist models for this job

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
30
Claude Opus 4.5 only
15
GLM-5 only
6
Like-for-like categories
2 / 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.

Math

Like-for-like
Claude Opus 4.5
57.9
#6/7
GLM-5
56.5
#7/7
Basis
Provisional lane · 4 vs 4 weighted rows
Reading
Claude Opus 4.5 leads

Multilingual

Like-for-like
Claude Opus 4.5
82.9
#2/12
GLM-5
48.7
#6/12
Basis
Provisional lane · 1 vs 1 weighted rows
Reading
Claude Opus 4.5 leads

Agentic

Directional only
Claude Opus 4.5
43.2
Estimated · #102/154
GLM-5
51.0
Estimated · #48/154
Basis
BenchAlign lane · 15 vs 11 public rows
Reading
Directional only

Coding

Directional only
Claude Opus 4.5
56.7
Estimated · #36/154
GLM-5
56.0
Estimated · #39/154
Basis
BenchAlign lane · 5 vs 6 public rows
Reading
Directional only

Knowledge

Directional only
Claude Opus 4.5
54.0
Estimated · #60/184
GLM-5
54.1
Estimated · #58/184
Basis
BenchAlign lane · 6 vs 6 public rows
Reading
Directional only

Instruction following

Directional only
Claude Opus 4.5
30.6
#114/124
GLM-5
87.2
#32/124
Basis
Provisional lane · 1 vs 0 weighted rows
Reading
Directional only

Reasoning

Not comparable
Claude Opus 4.5
72.1
Unranked · 4 rankable rows
GLM-5
52.1
Unranked · 4 rankable rows
Basis
Provisional lane · 1 vs 1 weighted rows
Reading
Not comparable

Multimodal

Not comparable
Claude Opus 4.5
23.5
#46/48
GLM-5
Not ranked
Basis
Provisional lane · 2 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.

  • HLE

    Knowledge

    Claude Opus 4.5: 30.8%GLM-5: 50.4%Normalized gap 19.6Shared source

  • FrontierMath v2 (Tiers 1-3)

    Math

    Claude Opus 4.5: 20.690%GLM-5: 16.434%Normalized gap 4.3Shared source

  • SWE Multilingual

    Coding

    Claude Opus 4.5: 77.5%GLM-5: 73.3%Normalized gap 4.2Shared source

  • SuperGPQA

    Knowledge

    Claude Opus 4.5: 70.6%GLM-5: 66.8%Normalized gap 3.8Shared source

  • MMLU-Pro

    Knowledge

    Claude Opus 4.5: 89.5%GLM-5: 85.7%Normalized gap 3.8Shared source

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

Claude Opus 4.5
$0.0175
Fits in one request
GLM-5
$0.0026
Fits in one request

GLM-5 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Claude Opus 4.5
$0.325
Fits in one request
GLM-5
$0.0596
Fits in one request

GLM-5 has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

Claude Opus 4.5
$1.35
Does not fit in one request
Cached input priced at the published list-input rate
GLM-5
$0.252
Does not fit in one request
Cached input priced at the published list-input rate

Claude Opus 4.5 does not fit this workload in one request. GLM-5 does not fit this workload in one request. Claude Opus 4.5 has no published cached-input rate, so cached tokens use its listed input rate. GLM-5 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.

Claude Opus 4.5

200K

GLM-5

200K

API model ID

Claude Opus 4.5

Not sourced

GLM-5

Not sourced

Cached-input rate

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

Claude Opus 4.5

Not published

GLM-5

Not published

Documented inputs

Claude Opus 4.5

Not sourced

GLM-5

Not sourced

Documented outputs

Claude Opus 4.5

Not sourced

GLM-5

Not sourced

Provider availability

Claude Opus 4.5

Not sourced

GLM-5

Not sourced

Reasoning profile

Claude Opus 4.5

Non-Reasoning

GLM-5

Non-Reasoning

Weight access

Claude Opus 4.5

Proprietary

GLM-5

Open Weight

License

Claude Opus 4.5

Proprietary

GLM-5

Open Weight

Release date

Claude Opus 4.5

2025-11-01

GLM-5

2026-03-01

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 has the higher public score estimate, 61.47 versus 58.42, but the 90% score intervals overlap.
Workload cost
Repository review: $0.325 vs $0.0596. Cache-heavy agent loop: $1.35 vs $0.252.
Context tradeoff
Both models list 200K.

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

Benchmark evidence

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

Browse raw public benchmark evidence51 rows

Agentic

  • Terminal-Bench 2.0

    Claude Opus 4.559.3%
    Source
    GLM-556.2%
    Source

    Claude Opus 4.5 leads this result

  • OSWorld-Verified

    Claude Opus 4.566.3%
    Source
    GLM-5—

    Not directly comparable

  • OSWorld

    Claude Opus 4.566.3%
    Source
    GLM-5—

    Not directly comparable

  • Claw-Eval

    Shared source
    Claude Opus 4.559.6%
    GLM-557.7%

    Claude Opus 4.5 leads this result

  • QwenClawBench

    Shared source
    Claude Opus 4.552.3%
    GLM-554.1%

    GLM-5 leads this result

  • τ³-bench results

    Shared source
    Claude Opus 4.570.2%
    GLM-565.6%

    Claude Opus 4.5 leads this result

  • VITA-Bench

    Claude Opus 4.523.3%
    Source
    GLM-5—

    Not directly comparable

  • DeepPlanning

    Shared source
    Claude Opus 4.526.4%
    GLM-514.6%

    Claude Opus 4.5 leads this result

  • Toolathlon

    Shared source
    Claude Opus 4.543.5%
    GLM-538%

    Claude Opus 4.5 leads this result

  • MCP Atlas

    Shared source
    Claude Opus 4.542.3%
    GLM-531.1%

    Claude Opus 4.5 leads this result

  • MCP-Tasks

    Shared source
    Claude Opus 4.571.8%
    GLM-560.8%

    Claude Opus 4.5 leads this result

  • WideResearch

    Shared source
    Claude Opus 4.576.4%
    GLM-569.8%

    Claude Opus 4.5 leads this result

  • CyberGym

    Shared source
    Claude Opus 4.550.6%
    GLM-543.2%

    Claude Opus 4.5 leads this result

  • Gert Labs

    Shared source
    Claude Opus 4.564.23%
    GLM-550.99%

    Claude Opus 4.5 leads this result

  • JobBench

    Claude Opus 4.532.3%
    Source
    GLM-5—

    Not directly comparable

Coding

  • SWE-bench Verified

    Claude Opus 4.580.9%
    Source
    GLM-577.8%
    Source

    Claude Opus 4.5 leads this result

  • LiveCodeBench v6

    Claude Opus 4.584.8%
    Source
    GLM-5—

    Not directly comparable

  • SWE-bench Pro

    Shared source
    Claude Opus 4.557.1%
    GLM-555.1%

    Claude Opus 4.5 leads this result

  • SWE Multilingual

    Shared source
    Claude Opus 4.577.5%
    GLM-573.3%

    Claude Opus 4.5 leads this result

  • NL2Repo

    Claude Opus 4.543.2%
    Source
    GLM-5—

    Not directly comparable

  • SWE-bench Verified*

    Claude Opus 4.5—
    GLM-572.8%
    Source

    Not directly comparable

  • SWE-Rebench

    Claude Opus 4.5—
    GLM-562.8%
    Source

    Not directly comparable

  • React Native Evals

    Claude Opus 4.5—
    GLM-574.8%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    Shared source
    Claude Opus 4.564.4%
    GLM-560.8%

    Claude Opus 4.5 leads this result

  • AI-Needle

    Shared source
    Claude Opus 4.574%
    GLM-563.3%

    Claude Opus 4.5 leads this result

Knowledge

  • GPQA

    Claude Opus 4.587%
    Source
    GLM-586%
    Source

    Claude Opus 4.5 leads this result

  • SuperGPQA

    Shared source
    Claude Opus 4.570.6%
    GLM-566.8%

    Claude Opus 4.5 leads this result

  • MMLU-Pro

    Shared source
    Claude Opus 4.589.5%
    GLM-585.7%

    Claude Opus 4.5 leads this result

  • MMLU-Redux

    Claude Opus 4.596.6%
    Source
    GLM-5—

    Not directly comparable

  • C-Eval

    Claude Opus 4.592.2%
    Source
    GLM-5—

    Not directly comparable

  • HLE

    Shared source
    Claude Opus 4.530.8%
    GLM-550.4%

    GLM-5 leads this result

  • GPQA-D

    Claude Opus 4.5—
    GLM-586.0%
    Source

    Not directly comparable

  • MMLU-Pro (Arcee)

    Claude Opus 4.5—
    GLM-585.8%
    Source

    Not directly comparable

Math

  • AIME26

    Shared source
    Claude Opus 4.595.1%
    GLM-595.8%

    GLM-5 leads this result

  • HMMT Feb 2025

    Shared source
    Claude Opus 4.592.9%
    GLM-597.5%

    GLM-5 leads this result

  • HMMT Nov 2025

    Shared source
    Claude Opus 4.593.3%
    GLM-596.9%

    GLM-5 leads this result

  • HMMT Feb 2026

    Shared source
    Claude Opus 4.585.3%
    GLM-586.4%

    GLM-5 leads this result

  • MMAnswerBench

    Shared source
    Claude Opus 4.584.0%
    GLM-582.5%

    Claude Opus 4.5 leads this result

  • FrontierMath v2 (Tiers 1-3)

    Shared source
    Claude Opus 4.520.690%
    GLM-516.434%

    Claude Opus 4.5 leads this result

  • FrontierMath v2 (Tier 4)

    Shared source
    Claude Opus 4.54.167%
    GLM-52.100%

    Claude Opus 4.5 leads this result

  • AIME25 (Arcee)

    Claude Opus 4.5—
    GLM-593.3%
    Source

    Not directly comparable

Multilingual

  • MMLU-ProX

    Shared source
    Claude Opus 4.585.7%
    GLM-583.1%

    Claude Opus 4.5 leads this result

  • NOVA-63

    Shared source
    Claude Opus 4.556.7%
    GLM-555.1%

    Claude Opus 4.5 leads this result

Multimodal

  • MMMU-Pro

    Claude Opus 4.570.6%
    Source
    GLM-5—

    Not directly comparable

  • MathVision

    Claude Opus 4.574.3%
    Source
    GLM-5—

    Not directly comparable

  • CharXiv

    Claude Opus 4.568.5%
    Source
    GLM-5—

    Not directly comparable

  • VideoMMMU

    Claude Opus 4.584.4%
    Source
    GLM-5—

    Not directly comparable

  • ScreenSpot Pro

    Claude Opus 4.545.7%
    Source
    GLM-5—

    Not directly comparable

  • V*

    Claude Opus 4.567.0%
    Source
    GLM-5—

    Not directly comparable

Instruction following

  • IFEval

    Shared source
    Claude Opus 4.590.9%
    GLM-592.6%

    GLM-5 leads this result

  • IFBench

    Claude Opus 4.558%
    Source
    GLM-5—

    Not directly comparable

Questions

Which is better, Claude Opus 4.5 or GLM-5?

GLM-5 has the higher public score estimate, 61.47 versus 58.42, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, Claude Opus 4.5 or GLM-5?

Claude Opus 4.5 scores higher for coding on the public lane, 56.7 to 56. Claude Opus 4.5 and GLM-5 are 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, Claude Opus 4.5 or GLM-5?

GLM-5 scores higher for agentic tasks on the public lane, 51 to 43.2. Claude Opus 4.5 and GLM-5 are 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, Claude Opus 4.5 or GLM-5?

For the stated presets, chat costs $0.0175 on Claude Opus 4.5 and $0.0026 on GLM-5; repository review costs $0.325 and $0.0596; the cache-heavy agent loop costs $1.35 and $0.252. Claude Opus 4.5 does not fit this workload in one request. GLM-5 does not fit this workload in one request. Claude Opus 4.5 has no published cached-input rate, so cached tokens use its listed input rate. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, Claude Opus 4.5 or GLM-5?

Both models list the same context window, 200K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated September 18, 2026

Watch Claude Opus 4.5 vs GLM-5

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

Read a sample issue

Join 2,000+ readers.