Skip to main content
BenchLM
Data

GLM-5.1 vs Qwen3.5 397B

Updated November 19, 2026. Rank says GLM-5.1 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInAPI/MCP

Decision reading

GLM-5.1 has the higher public point estimate, 52.2 versus 46.5. Their conditional score ranges overlap. These ranges do not establish rank confidence. 15 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A
Z.AI logo

Z.AI

52.2/100

Supported · Public rank #85

Conditional range 46.6–57.8

Model B
Alibaba logo

Alibaba

46.5/100

Supported · Public rank #107

Conditional range 41.5–51.5

API price per 1M tokens

GLM-5.1: $1.40 input / $4.40 output

Qwen3.5 397B: $0.60 input / $3.60 output

Context window

GLM-5.1: 203K

Qwen3.5 397B: 128K

Released

GLM-5.1: April 7, 2026

Qwen3.5 397B: February 16, 2026

Shared results
15
GLM-5.1 only
12
Qwen3.5 397B only
47
Like-for-like categories
1 / 8
Supported: GLM-5.1 and Qwen3.5 397B. Conditional ranges do not establish rank confidence.How the comparison works

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.

  • Agentic work

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

    GLM-5.1

    GLM-5.1 has the higher public agentic point estimate, 54.2 to 44.2, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

    Confidence: limited
  • Long documents

    Prompts that approach the documented context limit

    GLM-5.1

    GLM-5.1 has the larger documented context window.

    Confidence: documented
  • Chat turn cost

    1K fresh input + 500 output tokens

    Qwen3.5 397B

    Qwen3.5 397B 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

    Qwen3.5 397B

    Qwen3.5 397B 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

    Qwen3.5 397B is scored on Estimated evidence for coding, 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. GLM-5.1 does not fit this workload in one request. Qwen3.5 397B does not fit this workload in one request. GLM-5.1 has no published cached-input rate, so cached tokens use its listed input rate. Qwen3.5 397B 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.

51.3GLM-5.140.2Qwen3.5 397B

Directional only · BenchAlign v6.0

GLM-5.1 has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

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

Same basis rules as the category table below

Or describe your own job

The LLM Selector reads what you need a model for and returns a shortlist, ranked in the same order the job pages publish.

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy

What is actually comparable

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

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

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.

Bars run 0–100 on each benchmark’s normalized display scale

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v6.0 lane in every category, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Like-for-like
GLM-5.1
54.2
Supported · #79/177
Qwen3.5 397B
44.2
Supported · #115/177
Basis
BenchAlign v6.0 lane · 10 vs 14 public rows
Reading
GLM-5.1 leads · intervals overlap

Coding

Directional only
GLM-5.1
51.3
Supported · #76/177
Qwen3.5 397B
40.2
Estimated · #123/177
Basis
BenchAlign v6.0 lane · 8 vs 6 public rows
Reading
Directional only

Reasoning

Directional only
GLM-5.1
48.7
Supported · #83/160
Qwen3.5 397B
52.3
Estimated · #69/160
Basis
BenchAlign v6.0 lane · 0 vs 2 public rows
Reading
Directional only

Multimodal

Directional only
GLM-5.1
55.5
Estimated · #62/115
Qwen3.5 397B
51.5
Estimated · #67/115
Basis
BenchAlign v6.0 lane · 0 vs 22 public rows
Reading
Directional only

Knowledge

Directional only
GLM-5.1
47.4
Estimated · #91/156
Qwen3.5 397B
47.8
Estimated · #90/156
Basis
BenchAlign v6.0 lane · 4 vs 6 public rows
Reading
Directional only

Instruction following

Directional only
GLM-5.1
46.3
Estimated · #87/142
Qwen3.5 397B
45.9
Estimated · #81/142
Basis
BenchAlign v6.0 lane · 0 vs 2 public rows
Reading
Directional only

Math

Directional only
GLM-5.1
47.5
Estimated · #65/124
Qwen3.5 397B
43.3
Estimated · #77/124
Basis
BenchAlign v6.0 lane · 5 vs 5 public rows
Reading
Directional only

Multilingual

Not comparable
GLM-5.1
Not ranked
Qwen3.5 397B
50.9
Estimated · #43/86
Basis
BenchAlign v6.0 lane · 0 vs 5 public rows
Reading
Not comparable

Ranks count the models scored in each category’s BenchAlign v6.0 lane, so denominators differ by category. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

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

GLM-5.1
$0.0036
Fits in one request
Qwen3.5 397B
$0.0024
Fits in one request

Qwen3.5 397B has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

GLM-5.1
$0.0832
Fits in one request
Qwen3.5 397B
$0.0408
Fits in one request

Qwen3.5 397B has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

GLM-5.1
$0.352
Does not fit in one request
Cached input priced at the published list-input rate
Qwen3.5 397B
$0.168
Does not fit in one request
Cached input priced at the published list-input rate

GLM-5.1 does not fit this workload in one request. Qwen3.5 397B does not fit this workload in one request. GLM-5.1 has no published cached-input rate, so cached tokens use its listed input rate. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate.

Your workload

Enter a month of tokens to price it at both models’ listed API rates.

GLM-5.1
$22.80
per month
Qwen3.5 397B
$13.20
per month

Standard listed rates, no batch or volume discounts.

Cached input falls back to the list input rate only where a cached rate is unpublished

Price against score, among all priced models

Overall score against output price per 1M tokens, one model per family. Pricing as of November 18, 2026.

GLM-5.1Qwen3.5 397B

20406080100$0.10$0.50$1$5$10$50$100Output price, $ per 1M tokens (log)Overall score
  • 18 models score at least as high as GLM-5.1 for the same price or less. The nearest is GLM-5.2 (score 59.2, also $4.40 per 1M output tokens).
  • 16 cheaper models score at least as high as Qwen3.5 397B. The nearest is Gemini 3 Flash (score 53.1, $3.00 per 1M output tokens).

Gray dots are the other priced models. The dark step line is the efficiency frontier: for a model under it, another model costs no more and scores at least as high.

See Qwen3.5 397B on the full price vs performance chart

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.

GLM-5.1

203K

Qwen3.5 397B

128K

API model ID

GLM-5.1

Not sourced

Qwen3.5 397B

Not sourced

Cached-input rate

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

GLM-5.1

Not published

Qwen3.5 397B

Not published

Provider availability

GLM-5.1

Not sourced

Qwen3.5 397B

Not sourced

Reasoning profile

GLM-5.1

Reasoning

Qwen3.5 397B

Non-Reasoning

Weight access

GLM-5.1

Open Weight

Qwen3.5 397B

Open Weight

License

GLM-5.1

Open Weight

Qwen3.5 397B

Open Weight

Release date

GLM-5.1

2026-04-07

Qwen3.5 397B

2026-02-16

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-5.1 has the higher public point estimate, 52.2 versus 46.5. Their conditional score ranges overlap. These ranges do not establish rank confidence.
Workload cost
Repository review: $0.0832 vs $0.0408. Cache-heavy agent loop: $0.352 vs $0.168.
Context tradeoff
GLM-5.1 has the larger documented window (203K).
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, GLM-5.1 or Qwen3.5 397B?

GLM-5.1 has the higher public point estimate, 52.2 versus 46.5. Their conditional score ranges overlap. These ranges do not establish rank confidence. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, GLM-5.1 or Qwen3.5 397B?

GLM-5.1 scores higher for coding on the public lane, 51.3 to 40.2. Qwen3.5 397B 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, GLM-5.1 or Qwen3.5 397B?

GLM-5.1 has the higher public agentic tasks point estimate, 54.2 to 44.2, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

Which costs less, GLM-5.1 or Qwen3.5 397B?

For the stated presets, chat costs $0.0036 on GLM-5.1 and $0.0024 on Qwen3.5 397B; repository review costs $0.0832 and $0.0408; the cache-heavy agent loop costs $0.352 and $0.168. GLM-5.1 does not fit this workload in one request. Qwen3.5 397B does not fit this workload in one request. GLM-5.1 has no published cached-input rate, so cached tokens use its listed input rate. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, GLM-5.1 or Qwen3.5 397B?

GLM-5.1 has the larger documented context window: 203K, compared with 128K.

Self-host vs API cost

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

GLM-5.1
API / mo$4,350
Self-host / mo$27,962
Break-evenNot reached at 70% load
Qwen3.5 397B
API / mo$3,150
Self-host / mo$13,981
Break-even222M/day
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 evidence74 rows

Agentic

  • Terminal-Bench 2.0

    GLM-5.163.5%
    Source
    Qwen3.5 397B52.5%
    Source

    GLM-5.1 leads this result

  • BrowseComp

    GLM-5.168%
    Source
    Qwen3.5 397B69%
    Source

    Qwen3.5 397B leads this result

  • τ³-bench results

    GLM-5.170.6%
    Source
    Qwen3.5 397B68.4%
    Source

    GLM-5.1 leads this result

  • MCP Atlas

    GLM-5.171.8%
    Source
    Qwen3.5 397B46.1%
    Source

    GLM-5.1 leads this result

  • CyberGym

    GLM-5.168.7%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • GLM-5.162.3%
    Qwen3.5 397B56.8%

    GLM-5.1 leads this result

  • GLM-5.160.11%
    Qwen3.5 397B46.76%

    GLM-5.1 leads this result

  • ResearchClawBench

    Shared source
    GLM-5.118.2%
    Qwen3.5 397B14.2%

    GLM-5.1 leads this result

  • Terminal-Bench 2.1 (Vals)

    GLM-5.156.9%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • HLE w/ tools

    GLM-5.152.3%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • OSWorld-Verified

    GLM-5.1—
    Qwen3.5 397B62.2%
    Source

    Not directly comparable

  • QwenClawBench

    GLM-5.1—
    Qwen3.5 397B51.8%
    Source

    Not directly comparable

  • VITA-Bench

    GLM-5.1—
    Qwen3.5 397B43.7%
    Source

    Not directly comparable

  • DeepPlanning

    GLM-5.1—
    Qwen3.5 397B37.6%
    Source

    Not directly comparable

  • Toolathlon

    GLM-5.1—
    Qwen3.5 397B36.3%
    Source

    Not directly comparable

  • MCP-Tasks

    GLM-5.1—
    Qwen3.5 397B74.2%
    Source

    Not directly comparable

  • WideResearch

    GLM-5.1—
    Qwen3.5 397B74.0%
    Source

    Not directly comparable

Coding

  • SWE-bench Pro

    GLM-5.158.4%
    Source
    Qwen3.5 397B50.9%
    Source

    GLM-5.1 leads this result

  • NL2Repo

    GLM-5.142.7%
    Source
    Qwen3.5 397B32.2%
    Source

    GLM-5.1 leads this result

  • SWE-Rebench

    GLM-5.162.7%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • Terminal-Bench 2.0

    GLM-5.163.5%
    Source
    Qwen3.5 397B52.5%
    Source

    GLM-5.1 leads this result

  • Vibe Code Bench

    GLM-5.131.46%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • OpenHarmony Bench v1.0

    GLM-5.152.3%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • LiveCodeBench (Vals)

    GLM-5.181.4%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • SWE-bench (Vals)

    GLM-5.176.4%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • SWE-bench Verified

    GLM-5.1—
    Qwen3.5 397B76.2%
    Source

    Not directly comparable

  • LiveCodeBench v6

    GLM-5.1—
    Qwen3.5 397B83.6%
    Source

    Not directly comparable

  • SWE Multilingual

    GLM-5.1—
    Qwen3.5 397B69.3%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    GLM-5.1—
    Qwen3.5 397B63.2%
    Source

    Not directly comparable

  • AI-Needle

    GLM-5.1—
    Qwen3.5 397B68.7%
    Source

    Not directly comparable

Multimodal

  • MMMU

    GLM-5.1—
    Qwen3.5 397B85.0%
    Source

    Not directly comparable

  • MMMU-Pro

    GLM-5.1—
    Qwen3.5 397B79%
    Source

    Not directly comparable

  • RealWorldQA

    GLM-5.1—
    Qwen3.5 397B83.9%
    Source

    Not directly comparable

  • OmniDocBench 1.5

    GLM-5.1—
    Qwen3.5 397B90.8%
    Source

    Not directly comparable

  • Video-MME (with subtitle)

    GLM-5.1—
    Qwen3.5 397B87.5%
    Source

    Not directly comparable

  • Video-MME (w/o subtitle)

    GLM-5.1—
    Qwen3.5 397B83.7%
    Source

    Not directly comparable

  • MathVision

    GLM-5.1—
    Qwen3.5 397B88.6%
    Source

    Not directly comparable

  • We-Math

    GLM-5.1—
    Qwen3.5 397B87.9%
    Source

    Not directly comparable

  • DynaMath

    GLM-5.1—
    Qwen3.5 397B86.3%
    Source

    Not directly comparable

  • SimpleVQA

    GLM-5.1—
    Qwen3.5 397B67.1%
    Source

    Not directly comparable

  • CharXiv

    GLM-5.1—
    Qwen3.5 397B80.8%
    Source

    Not directly comparable

  • MMLongBench-Doc

    GLM-5.1—
    Qwen3.5 397B61.5%
    Source

    Not directly comparable

  • CC-OCR

    GLM-5.1—
    Qwen3.5 397B82.0%
    Source

    Not directly comparable

  • AI2D_TEST

    GLM-5.1—
    Qwen3.5 397B93.9%
    Source

    Not directly comparable

  • CountBench

    GLM-5.1—
    Qwen3.5 397B97.2%
    Source

    Not directly comparable

  • RefCOCO (avg)

    GLM-5.1—
    Qwen3.5 397B92.3%
    Source

    Not directly comparable

  • ODINW13

    GLM-5.1—
    Qwen3.5 397B47.0%
    Source

    Not directly comparable

  • ERQA

    GLM-5.1—
    Qwen3.5 397B67.5%
    Source

    Not directly comparable

  • VideoMMMU

    GLM-5.1—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • MLVU (M-Avg)

    GLM-5.1—
    Qwen3.5 397B86.7%
    Source

    Not directly comparable

  • ScreenSpot Pro

    GLM-5.1—
    Qwen3.5 397B65.6%
    Source

    Not directly comparable

  • V*

    GLM-5.1—
    Qwen3.5 397B95.8%
    Source

    Not directly comparable

Knowledge

  • GPQA-D

    GLM-5.186.2%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • HLE

    GLM-5.131%
    Source
    Qwen3.5 397B28.7%
    Source

    GLM-5.1 leads this result

  • GPQA Diamond (Vals)

    GLM-5.184.5%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • MMLU-Pro (Vals)

    GLM-5.186.9%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • GPQA

    GLM-5.1—
    Qwen3.5 397B88.4%
    Source

    Not directly comparable

  • SuperGPQA

    GLM-5.1—
    Qwen3.5 397B70.4%
    Source

    Not directly comparable

  • MMLU-Pro

    GLM-5.1—
    Qwen3.5 397B87.8%
    Source

    Not directly comparable

  • MMLU-Redux

    GLM-5.1—
    Qwen3.5 397B94.9%
    Source

    Not directly comparable

  • C-Eval

    GLM-5.1—
    Qwen3.5 397B93%
    Source

    Not directly comparable

Multilingual

  • MMLU-ProX

    GLM-5.1—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • NOVA-63

    GLM-5.1—
    Qwen3.5 397B59.1%
    Source

    Not directly comparable

  • INCLUDE

    GLM-5.1—
    Qwen3.5 397B85.6%
    Source

    Not directly comparable

  • PolyMath

    GLM-5.1—
    Qwen3.5 397B73.3%
    Source

    Not directly comparable

  • MAXIFE

    GLM-5.1—
    Qwen3.5 397B88.2%
    Source

    Not directly comparable

Instruction following

  • IFEval

    GLM-5.1—
    Qwen3.5 397B92.6%
    Source

    Not directly comparable

  • IFBench

    GLM-5.1—
    Qwen3.5 397B76.5%
    Source

    Not directly comparable

Math

  • AIME26

    GLM-5.195.3%
    Source
    Qwen3.5 397B93.3%
    Source

    GLM-5.1 leads this result

  • HMMT Nov 2025

    GLM-5.194.0%
    Source
    Qwen3.5 397B92.7%
    Source

    GLM-5.1 leads this result

  • HMMT Feb 2026

    GLM-5.182.6%
    Source
    Qwen3.5 397B87.9%
    Source

    Qwen3.5 397B leads this result

  • MMAnswerBench

    GLM-5.183.8%
    Source
    Qwen3.5 397B80.9%
    Source

    GLM-5.1 leads this result

  • FrontierMath v2 (Tiers 1-3)

    GLM-5.136.842%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • HMMT Feb 2025

    GLM-5.1—
    Qwen3.5 397B94.8%
    Source

    Not directly comparable

74 public results · 15 shared

Watch GLM-5.1 vs Qwen3.5 397B

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

Read a sample issue

Join 7,000+ subscribers.

Last updated November 19, 2026