Skip to main content
BenchLM
Advertise

Model comparison

GLM-5-Turbo vs Kimi K2.5

Updated July 29, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

GLM-5-Turbo

Z.AI

65.9/100

Supported · Public rank #29

90% interval 56.1–75.7

Kimi K2.5

Moonshot AI

58.8/100

Supported · Public rank #60

90% interval 51.3–66.3

GLM-5-Turbo has the higher public score estimate, 65.92 versus 58.78, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

1 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

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

    Kimi K2.5

    Kimi K2.5 has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    Kimi K2.5

    Kimi K2.5 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

    Kimi K2.5

    Kimi K2.5 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

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    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-Turbo does not fit this workload in one request. GLM-5-Turbo has no published cached-input rate, so cached tokens use its listed input rate. Kimi K2.5 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
GLM-5-Turbo only
0
Kimi K2.5 only
44
Like-for-like categories
0 / 8

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Agentic

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
55.0
Weighted basis
0 vs 2 rows
Reading
Not comparable

Coding

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
59.4
Weighted basis
0 vs 4 rows
Reading
Not comparable

Reasoning

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
61.0
Weighted basis
0 vs 1 rows
Reading
Not comparable

Knowledge

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
56.9
Weighted basis
0 vs 4 rows
Reading
Not comparable

Math

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
60.6
Weighted basis
0 vs 4 rows
Reading
Not comparable

Multilingual

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
82.3
Weighted basis
0 vs 1 rows
Reading
Not comparable

Multimodal

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
78.5
Weighted basis
0 vs 1 rows
Reading
Not comparable

Instruction following

Not comparable
GLM-5-Turbo
Not measured
Kimi K2.5
93.9
Weighted basis
0 vs 1 rows
Reading
Not comparable

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.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

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-Turbo
$0.0032
Fits in one request
Kimi K2.5
$0.0021
Fits in one request

Kimi K2.5 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

GLM-5-Turbo
$0.072
Fits in one request
Kimi K2.5
$0.039
Fits in one request

Kimi K2.5 has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

GLM-5-Turbo
$0.304
Does not fit in one request
Cached input priced at the published list-input rate
Kimi K2.5
$0.162
Fits in one request
Cached input priced at the published list-input rate

GLM-5-Turbo does not fit this workload in one request. GLM-5-Turbo has no published cached-input rate, so cached tokens use its listed input rate. Kimi K2.5 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.

GLM-5-Turbo

200K

Kimi K2.5

256K

API model ID

GLM-5-Turbo

Not sourced

Kimi K2.5

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-Turbo

Not published

Kimi K2.5

Not published

Documented inputs

GLM-5-Turbo

Not sourced

Kimi K2.5

Not sourced

Documented outputs

GLM-5-Turbo

Not sourced

Kimi K2.5

Not sourced

Provider availability

GLM-5-Turbo

Not sourced

Kimi K2.5

Not sourced

Reasoning profile

GLM-5-Turbo

Reasoning

Kimi K2.5

Non-Reasoning

Weight access

GLM-5-Turbo

Proprietary

Kimi K2.5

Open Weight

License

GLM-5-Turbo

Proprietary

Kimi K2.5

Open Weight

Release date

GLM-5-Turbo

2026-03-01

Kimi K2.5

2026-02-01

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-Turbo has the higher public score estimate, 65.92 versus 58.78, but the 90% score intervals overlap.
Workload cost
Repository review: $0.072 vs $0.039. Cache-heavy agent loop: $0.304 vs $0.162.
Context tradeoff
Kimi K2.5 has the larger documented window (256K).

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.

GLM-5-Turbo
API / mo$3,900
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Kimi K2.5
API / mo$2,700
Self-host / mo$5,221
Break-even132M/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 evidence45 rows

Agentic

  • Claw-Eval

    Shared source
    GLM-5-Turbo55.8%
    Kimi K2.552.3%

    GLM-5-Turbo leads this result

  • Terminal-Bench 2.0

    GLM-5-Turbo—
    Kimi K2.550.8%
    Source

    Not directly comparable

  • BrowseComp

    GLM-5-Turbo—
    Kimi K2.560.6%
    Source

    Not directly comparable

  • QwenClawBench

    GLM-5-Turbo—
    Kimi K2.554.3%
    Source

    Not directly comparable

  • τ³-bench results

    GLM-5-Turbo—
    Kimi K2.565.7%
    Source

    Not directly comparable

  • DeepSearchQA

    GLM-5-Turbo—
    Kimi K2.577.1%
    Source

    Not directly comparable

  • DeepPlanning

    GLM-5-Turbo—
    Kimi K2.514.4%
    Source

    Not directly comparable

  • Toolathlon

    GLM-5-Turbo—
    Kimi K2.527.8%
    Source

    Not directly comparable

  • MCP Atlas

    GLM-5-Turbo—
    Kimi K2.529.5%
    Source

    Not directly comparable

  • MCP-Tasks

    GLM-5-Turbo—
    Kimi K2.559.1%
    Source

    Not directly comparable

  • WideResearch

    GLM-5-Turbo—
    Kimi K2.572.7%
    Source

    Not directly comparable

  • Gert Labs

    GLM-5-Turbo—
    Kimi K2.545.88%
    Source

    Not directly comparable

  • ResearchClawBench

    GLM-5-Turbo—
    Kimi K2.514.0%
    Source

    Not directly comparable

  • JobBench

    GLM-5-Turbo—
    Kimi K2.58.7%
    Source

    Not directly comparable

Coding

  • SWE-bench Verified

    GLM-5-Turbo—
    Kimi K2.576.8%
    Source

    Not directly comparable

  • SWE-bench Verified*

    GLM-5-Turbo—
    Kimi K2.570.8%
    Source

    Not directly comparable

  • LiveCodeBench v6

    GLM-5-Turbo—
    Kimi K2.585.0%
    Source

    Not directly comparable

  • SWE-bench Pro

    GLM-5-Turbo—
    Kimi K2.550.7%
    Source

    Not directly comparable

  • SWE Multilingual

    GLM-5-Turbo—
    Kimi K2.573%
    Source

    Not directly comparable

  • SWE-Rebench

    GLM-5-Turbo—
    Kimi K2.558.5%
    Source

    Not directly comparable

  • React Native Evals

    GLM-5-Turbo—
    Kimi K2.577.2%
    Source

    Not directly comparable

  • SciCode

    GLM-5-Turbo—
    Kimi K2.548.7%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    GLM-5-Turbo—
    Kimi K2.561%
    Source

    Not directly comparable

Knowledge

  • GPQA

    GLM-5-Turbo—
    Kimi K2.587.6%
    Source

    Not directly comparable

  • GPQA-D

    GLM-5-Turbo—
    Kimi K2.587.6%
    Source

    Not directly comparable

  • SuperGPQA

    GLM-5-Turbo—
    Kimi K2.569.2%
    Source

    Not directly comparable

  • MMLU-Pro

    GLM-5-Turbo—
    Kimi K2.587.1%
    Source

    Not directly comparable

  • MMLU-Pro (Arcee)

    GLM-5-Turbo—
    Kimi K2.587.1%
    Source

    Not directly comparable

  • HLE

    GLM-5-Turbo—
    Kimi K2.530.1%
    Source

    Not directly comparable

Math

  • AIME 2025

    GLM-5-Turbo—
    Kimi K2.596.1%
    Source

    Not directly comparable

  • AIME26

    GLM-5-Turbo—
    Kimi K2.595.8%
    Source

    Not directly comparable

  • AIME25 (Arcee)

    GLM-5-Turbo—
    Kimi K2.596.3%
    Source

    Not directly comparable

  • HMMT Feb 2025

    GLM-5-Turbo—
    Kimi K2.595.4%
    Source

    Not directly comparable

  • HMMT Nov 2025

    GLM-5-Turbo—
    Kimi K2.591.1%
    Source

    Not directly comparable

  • HMMT Feb 2026

    GLM-5-Turbo—
    Kimi K2.587.1%
    Source

    Not directly comparable

  • MMAnswerBench

    GLM-5-Turbo—
    Kimi K2.581.8%
    Source

    Not directly comparable

  • FrontierMath v2 (Tiers 1-3)

    GLM-5-Turbo—
    Kimi K2.527.900%
    Source

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    GLM-5-Turbo—
    Kimi K2.54.200%
    Source

    Not directly comparable

Multilingual

  • MMLU-ProX

    GLM-5-Turbo—
    Kimi K2.582.3%
    Source

    Not directly comparable

  • NOVA-63

    GLM-5-Turbo—
    Kimi K2.556.0%
    Source

    Not directly comparable

Multimodal

  • MMMU-Pro

    GLM-5-Turbo—
    Kimi K2.578.5%
    Source

    Not directly comparable

  • Video-MME

    GLM-5-Turbo—
    Kimi K2.587.4%
    Source

    Not directly comparable

  • MMVU

    GLM-5-Turbo—
    Kimi K2.580.4%
    Source

    Not directly comparable

  • VideoMMMU

    GLM-5-Turbo—
    Kimi K2.586.6%
    Source

    Not directly comparable

Instruction following

  • IFEval

    GLM-5-Turbo—
    Kimi K2.593.9%
    Source

    Not directly comparable

Frequently asked questions

Which is better, GLM-5-Turbo or Kimi K2.5?

GLM-5-Turbo has the higher public score estimate, 65.92 versus 58.78, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, GLM-5-Turbo or Kimi K2.5?

The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.

Which is better for agentic tasks, GLM-5-Turbo or Kimi K2.5?

The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.

Which costs less, GLM-5-Turbo or Kimi K2.5?

For the stated presets, chat costs $0.0032 on GLM-5-Turbo and $0.0021 on Kimi K2.5; repository review costs $0.072 and $0.039; the cache-heavy agent loop costs $0.304 and $0.162. GLM-5-Turbo does not fit this workload in one request. GLM-5-Turbo has no published cached-input rate, so cached tokens use its listed input rate. Kimi K2.5 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, GLM-5-Turbo or Kimi K2.5?

Kimi K2.5 has the larger documented context window: 256K, compared with 200K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated July 29, 2026

Watch GLM-5-Turbo vs Kimi K2.5

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

Read a sample issue

Join 2,000+ readers.