Long documents
Prompts that approach the documented context limit
Kimi K3
Kimi K3 has the larger documented context window.
Confidence: documented
GLM-5 vs Kimi K3
Updated October 2, 2026. Rank says Kimi K3 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.
Both of these models will change. Get the price, version and retirement notices for the pair, each with its source. Follow model changes
Share or export
Decision reading
Kimi K3 has the higher public score, 72.14 versus 55.06, and the 90% score intervals do not overlap. 4 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 4
- GLM-5 only
- 32
- Kimi K3 only
- 44
- Like-for-like categories
- 0 / 8
Supported: GLM-5 and Kimi K3How 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.
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-ratesCoding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
GLM-5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limitedAgentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
GLM-5 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limitedCache-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 does not fit this workload in one request. 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.
40.4GLM-561.4Kimi K3
Directional only · BenchAlign v5.8
Kimi K3 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
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
32 GLM-5 only
4 Shared
44 Kimi K3 only
3 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.
GPQAKnowledge
Normalized gap 7.5GLM-5:86%Kimi K3:93.5%HLEKnowledge
Normalized gap 5.6GLM-5:50.4%Kimi K3:56%
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 v5.8 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- GLM-5
- 38.8
- Estimated · #62/119
- Kimi K3
- 68.1
- Supported · #8/119
- Basis
- BenchAlign v5.8 lane · 11 vs 12 public rows
- Reading
- Directional only
Coding
Directional only- GLM-5
- 40.4
- Estimated · #63/144
- Kimi K3
- 61.4
- Supported · #18/144
- Basis
- BenchAlign v5.8 lane · 6 vs 14 public rows
- Reading
- Directional only
Knowledge
Directional only- GLM-5
- 48.5
- Estimated · #72/171
- Kimi K3
- 67.9
- Supported · #18/171
- Basis
- BenchAlign v5.8 lane · 6 vs 6 public rows
- Reading
- Directional only
Reasoning
Not comparable- GLM-5
- 53.3
- Unranked · 4 rankable rows
- Kimi K3
- 65.8
- #18/27
- Basis
- Provisional lane · 1 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- GLM-5
- Not ranked
- Kimi K3
- 89.4
- #1/49
- Basis
- Provisional lane · 0 vs 3 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- GLM-5
- 89.5
- #6/16
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- GLM-5
- 87.2
- #31/125
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- GLM-5
- 56.5
- #7/7
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 4 vs 0 weighted rows
- Reading
- Not comparable
| Category | GLM-5 | Kimi K3 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 38.8Estimated · #62/119 | 68.1Supported · #8/119 | Directional onlyBenchAlign v5.8 lane · 11 vs 12 public rows | Directional only |
| Coding | 40.4Estimated · #63/144 | 61.4Supported · #18/144 | Directional onlyBenchAlign v5.8 lane · 6 vs 14 public rows | Directional only |
| Knowledge | 48.5Estimated · #72/171 | 67.9Supported · #18/171 | Directional onlyBenchAlign v5.8 lane · 6 vs 6 public rows | Directional only |
| Reasoning | 53.3Unranked · 4 rankable rows | 65.8#18/27 | Not comparableProvisional lane · 1 vs 1 weighted rows | Not comparable |
| Multimodal | Not ranked | 89.4#1/49 | Not comparableProvisional lane · 0 vs 3 weighted rows | Not comparable |
| Multilingual | 89.5#6/16 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Instruction following | 87.2#31/125 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 56.5#7/7 | Not ranked | Not comparableProvisional lane · 4 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. 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
- $0.0026
- Fits in one request
- Kimi K3
- $0.0105
- 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
- GLM-5
- $0.0596
- Fits in one request
- Kimi K3
- $0.195
- 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
- GLM-5
- $0.252
- Does not fit in one request
- Cached input priced at the published list-input rate
- Kimi K3
- $0.27
- Fits in one request
GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationGLM-5Kimi K3
Context window
Maximum documented context; output-token limits may be lower.
GLM-5
200K
Kimi K3
1.05M
API model ID
GLM-5
Not sourced
Kimi K3
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
Not published
Kimi K3
$0.3 per 1M cached input tokens
Documented inputs
GLM-5
Not sourced
Kimi K3
Not sourced
Documented outputs
GLM-5
Not sourced
Kimi K3
Not sourced
Provider availability
GLM-5
Not sourced
Kimi K3
Not sourced
Reasoning profile
GLM-5
Non-Reasoning
Kimi K3
Reasoning
Weight access
GLM-5
Open Weight
Kimi K3
Pending
License
GLM-5
Open Weight
Kimi K3
Pending
Release date
GLM-5
2026-03-01
Kimi K3
2026-07-16
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
- Kimi K3 has the higher public score, 72.14 versus 55.06, and the 90% score intervals do not overlap.
- Workload cost
- Repository review: $0.0596 vs $0.195. Cache-heavy agent loop: $0.252 vs $0.27.
- Context tradeoff
- Kimi K3 has the larger documented window (1.05M).
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, GLM-5 or Kimi K3?
Kimi K3 has the higher public score, 72.14 versus 55.06, 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, GLM-5 or Kimi K3?
Kimi K3 scores higher for coding on the public lane, 61.4 to 40.4. GLM-5 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 or Kimi K3?
Kimi K3 scores higher for agentic tasks on the public lane, 68.1 to 38.8. GLM-5 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, GLM-5 or Kimi K3?
For the stated presets, chat costs $0.0026 on GLM-5 and $0.0105 on Kimi K3; repository review costs $0.0596 and $0.195; the cache-heavy agent loop costs $0.252 and $0.27. GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, GLM-5 or Kimi K3?
Kimi K3 has the larger documented context window: 1.05M, compared with 200K.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence80 rows
Agentic
Terminal-Bench 2.0
GLM-556.2%SourceKimi K3—Not directly comparable
Claw-Eval
GLM-557.7%SourceKimi K3—Not directly comparable
QwenClawBench
GLM-554.1%SourceKimi K3—Not directly comparable
τ³-bench results
GLM-565.6%SourceKimi K3—Not directly comparable
DeepPlanning
GLM-514.6%SourceKimi K3—Not directly comparable
Toolathlon
GLM-538%SourceKimi K3—Not directly comparable
MCP Atlas
GLM-531.1%SourceKimi K384.2%SourceKimi K3 leads this result
MCP-Tasks
GLM-560.8%SourceKimi K3—Not directly comparable
WideResearch
GLM-569.8%SourceKimi K3—Not directly comparable
CyberGym
GLM-543.2%SourceKimi K3—Not directly comparable
Gert Labs
GLM-550.99%SourceKimi K3—Not directly comparable
Terminal-Bench 2.1
GLM-5—Kimi K388.3%SourceNot directly comparable
BrowseComp
GLM-5—Kimi K391.2%SourceNot directly comparable
DeepSearchQA
GLM-5—Kimi K395.0%SourceNot directly comparable
Toolathlon-Verified
GLM-5—Kimi K373.2%SourceNot directly comparable
AutomationBench
GLM-5—Kimi K330.8%SourceNot directly comparable
JobBench
GLM-5—Kimi K352.9%SourceNot directly comparable
APEX-Agents
GLM-5—Kimi K337.6%SourceNot directly comparable
SpreadsheetBench 2
GLM-5—Kimi K334.8%SourceNot directly comparable
DECK-Bench
GLM-5—Kimi K373.5%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
GLM-5—Kimi K380.9%SourceNot directly comparable
ApprenticeBench
GLM-5—Kimi K318%SourceNot directly comparable
Coding
SWE-bench Verified
GLM-577.8%SourceKimi K3—Not directly comparable
SWE-bench Verified*
GLM-572.8%SourceKimi K3—Not directly comparable
SWE-bench Pro
GLM-555.1%SourceKimi K3—Not directly comparable
SWE Multilingual
GLM-573.3%SourceKimi K3—Not directly comparable
SWE-Rebench
GLM-562.8%SourceKimi K3—Not directly comparable
React Native Evals
GLM-574.8%SourceKimi K3—Not directly comparable
DeepSWE
GLM-5—Kimi K367.5%SourceNot directly comparable
CursorBench 3.2
GLM-5—Kimi K360.8%SourceNot directly comparable
FrontierSWE
GLM-5—Kimi K381.2%SourceNot directly comparable
ProgramBench
GLM-5—Kimi K377.8%SourceNot directly comparable
Kimi Code Bench v2
GLM-5—Kimi K372.9%SourceNot directly comparable
sweMarathon
GLM-5—Kimi K342%SourceNot directly comparable
PostTrain Bench
GLM-5—Kimi K336.6%SourceNot directly comparable
MLS-Bench Lite
GLM-5—Kimi K348.3%SourceNot directly comparable
VulcanBench v3
GLM-5—Kimi K373.7%SourceNot directly comparable
OpenHarmony Bench
GLM-5—Kimi K357.3%SourceNot directly comparable
FrontierSWE v2
GLM-5—Kimi K325.9%SourceNot directly comparable
LiveCodeBench (Vals)
GLM-5—Kimi K387.2%SourceNot directly comparable
SWE-bench (Vals)
GLM-5—Kimi K393.4%SourceNot directly comparable
PostTrainBench v1.1
GLM-5—Kimi K332.0%SourceNot directly comparable
Reasoning
Multimodal
OfficeQA Pro
GLM-5—Kimi K363.3%SourceNot directly comparable
MMMU-Pro
GLM-5—Kimi K381.6%SourceNot directly comparable
MMMU-Pro w/ Python
GLM-5—Kimi K383.4%SourceNot directly comparable
CharXiv w/o tools
GLM-5—Kimi K384.8%SourceNot directly comparable
CharXiv
GLM-5—Kimi K391.3%SourceNot directly comparable
MathVision
GLM-5—Kimi K394.3%SourceNot directly comparable
MathVision w/ Python
GLM-5—Kimi K397.8%SourceNot directly comparable
BabyVision w/ Python
GLM-5—Kimi K385.7%SourceNot directly comparable
ZeroBench
GLM-5—Kimi K323.0%SourceNot directly comparable
ZeroBench w/ Python
GLM-5—Kimi K341.0%SourceNot directly comparable
WorldVQA ForceAnswer
GLM-5—Kimi K351.0%SourceNot directly comparable
OmniDocBench
GLM-5—Kimi K391.1%SourceNot directly comparable
PerceptionBench
GLM-5—Kimi K358.5%SourceNot directly comparable
Knowledge
GPQA
GLM-586%SourceKimi K393.5%SourceKimi K3 leads this result
GPQA-D
GLM-586.0%SourceKimi K393.5%SourceKimi K3 leads this result
SuperGPQA
GLM-566.8%SourceKimi K3—Not directly comparable
MMLU-Pro
GLM-585.7%SourceKimi K3—Not directly comparable
MMLU-Pro (Arcee)
GLM-585.8%SourceKimi K3—Not directly comparable
HLE
GLM-550.4%SourceKimi K356%SourceKimi K3 leads this result
HLE w/o tools
GLM-5—Kimi K343.5%SourceNot directly comparable
GPQA Diamond (Vals)
GLM-5—Kimi K392.9%SourceNot directly comparable
MMLU-Pro (Vals)
GLM-5—Kimi K388.0%SourceNot directly comparable
Multilingual
Instruction following
Math
AIME26
GLM-595.8%SourceKimi K3—Not directly comparable
AIME25 (Arcee)
GLM-593.3%SourceKimi K3—Not directly comparable
HMMT Feb 2025
GLM-597.5%SourceKimi K3—Not directly comparable
HMMT Nov 2025
GLM-596.9%SourceKimi K3—Not directly comparable
HMMT Feb 2026
GLM-586.4%SourceKimi K3—Not directly comparable
MMAnswerBench
GLM-582.5%SourceKimi K3—Not directly comparable
FrontierMath v2 (Tiers 1-3)
GLM-516.434%SourceKimi K3—Not directly comparable
FrontierMath v2 (Tier 4)
GLM-52.100%SourceKimi K3—Not directly comparable
80 public results · 4 shared
Watch GLM-5 vs Kimi K3
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.
Last updated October 2, 2026