Agentic
Like-for-like- GLM-5.2
- 58.5
- Supported · #29/153
- Kimi K2.6
- 46.6
- Supported · #77/153
- Basis
- BenchAlign lane · 6 vs 12 public rows
- Reading
- GLM-5.2 leads · intervals overlap
GLM-5.2 vs Kimi K2.6
Updated September 14, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
GLM-5.2 has the higher public score estimate, 68.19 versus 65.46, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
17 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
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.
Coding work
Code generation, repair, and software-engineering tasks
GLM-5.2
GLM-5.2 leads on the public coding lane, 61 to 50.8, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
GLM-5.2
GLM-5.2 leads on the public agentic lane, 58.5 to 46.6, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Long documents
Prompts that approach the documented context limit
GLM-5.2
GLM-5.2 has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Chat turn cost
1K fresh input + 500 output tokens
Kimi K2.6
Kimi K2.6 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Kimi K2.6
Kimi K2.6 has the lower estimated token cost for this stated workload. GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate. Kimi K2.6 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
Repository review cost
50K fresh input + 3K output tokens
Kimi K2.6
Kimi K2.6 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
8 GLM-5.2 only
17 Shared
20 Kimi K2.6 only
- Shared results
- 17
- GLM-5.2 only
- 8
- Kimi K2.6 only
- 20
- Like-for-like categories
- 3 / 8
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.
Coding
Like-for-like- GLM-5.2
- 61.0
- Supported · #19/152
- Kimi K2.6
- 50.8
- Supported · #53/152
- Basis
- BenchAlign lane · 8 vs 10 public rows
- Reading
- GLM-5.2 leads · intervals overlap
Knowledge
Like-for-like- GLM-5.2
- 60.7
- Supported · #35/183
- Kimi K2.6
- 61.4
- Supported · #32/183
- Basis
- BenchAlign lane · 6 vs 5 public rows
- Reading
- Kimi K2.6 leads · intervals overlap
Reasoning
Not comparable- GLM-5.2
- 74.8
- Unranked · 2 rankable rows
- Kimi K2.6
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- GLM-5.2
- 80.7
- Unranked · 4 rankable rows
- Kimi K2.6
- 71.3
- #1/7
- Basis
- Provisional lane · 2 vs 4 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- GLM-5.2
- Not ranked
- Kimi K2.6
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- GLM-5.2
- Not ranked
- Kimi K2.6
- 64.0
- #26/48
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- GLM-5.2
- 89.8
- #22/123
- Kimi K2.6
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | GLM-5.2 | Kimi K2.6 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 58.5Supported · #29/153 | 46.6Supported · #77/153 | Like-for-likeBenchAlign lane · 6 vs 12 public rows | GLM-5.2 leads · intervals overlap |
| Coding | 61.0Supported · #19/152 | 50.8Supported · #53/152 | Like-for-likeBenchAlign lane · 8 vs 10 public rows | GLM-5.2 leads · intervals overlap |
| Knowledge | 60.7Supported · #35/183 | 61.4Supported · #32/183 | Like-for-likeBenchAlign lane · 6 vs 5 public rows | Kimi K2.6 leads · intervals overlap |
| Reasoning | 74.8Unranked · 2 rankable rows | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 80.7Unranked · 4 rankable rows | 71.3#1/7 | Not comparableProvisional lane · 2 vs 4 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | Not ranked | 64.0#26/48 | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Instruction following | 89.8#22/123 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | 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
LiveCodeBench (Vals)
Coding
Terminal-Bench 2.0
Agentic
SWE-bench Pro
Coding
AIME26
Math
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-5.2
- $0.0036
- Fits in one request
- Kimi K2.6
- $0.00295
- Fits in one request
Kimi K2.6 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- GLM-5.2
- $0.0832
- Fits in one request
- Kimi K2.6
- $0.0595
- Fits in one request
Kimi K2.6 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-5.2
- $0.352
- Fits in one request
- Cached input priced at the published list-input rate
- Kimi K2.6
- $0.249
- Fits in one request
- Cached input priced at the published list-input rate
Kimi K2.6 has the lower modeled cost
GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate. Kimi K2.6 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.
SpecificationGLM-5.2Kimi K2.6
Context window
Maximum documented context; output-token limits may be lower.
GLM-5.2
1M
Kimi K2.6
256K
API model ID
GLM-5.2
Not sourced
Kimi K2.6
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GLM-5.2
Not published
Kimi K2.6
Not published
Documented inputs
GLM-5.2
Not sourced
Kimi K2.6
Not sourced
Documented outputs
GLM-5.2
Not sourced
Kimi K2.6
Not sourced
Provider availability
GLM-5.2
Not sourced
Kimi K2.6
Not sourced
Reasoning profile
GLM-5.2
Reasoning
Kimi K2.6
Reasoning
Weight access
GLM-5.2
Open Weight
Kimi K2.6
Open Weight
License
GLM-5.2
Open Weight
Kimi K2.6
Open Weight
Release date
GLM-5.2
2026-06-16
Kimi K2.6
2026-04-20
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 estimate, 68.19 versus 65.46, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.0832 vs $0.0595. Cache-heavy agent loop: $0.352 vs $0.249.
- 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.
GLM-5.2
API / mo$4,350
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Kimi K2.6
API / mo$3,713
Self-host / mo$18,221
Break-even326M/day
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
Terminal-Bench 3.0
GLM-5.24.6%SourceKimi K2.6—Not directly comparable
Terminal-Bench 2.0
GLM-5.281%SourceKimi K2.666.7%SourceGLM-5.2 leads this result
MCP Atlas
GLM-5.276.8%SourceKimi K2.655.9%SourceGLM-5.2 leads this result
Toolathlon
GLM-5.248.2%SourceKimi K2.650%SourceKimi K2.6 leads this result
ResearchClawBench
Shared sourceGLM-5.220.7%Kimi K2.618.0%GLM-5.2 leads this result
Terminal-Bench 2.1 (Vals)
GLM-5.267.8%SourceKimi K2.653.6%SourceGLM-5.2 leads this result
BrowseComp
GLM-5.2—Kimi K2.683.2%SourceNot directly comparable
OSWorld-Verified
GLM-5.2—Kimi K2.673.1%SourceNot directly comparable
Claw-Eval
GLM-5.2—Kimi K2.662.3%SourceNot directly comparable
DeepSearchQA
GLM-5.2—Kimi K2.692.5%SourceNot directly comparable
WideResearch
GLM-5.2—Kimi K2.680.8%SourceNot directly comparable
Gert Labs
GLM-5.2—Kimi K2.656.82%SourceNot directly comparable
OSWorld 2.0
GLM-5.2—Kimi K2.64.6%SourceNot directly comparable
Coding
SWE-bench Pro
GLM-5.262.1%SourceKimi K2.658.6%SourceGLM-5.2 leads this result
NL2Repo
GLM-5.248.9%SourceKimi K2.6—Not directly comparable
Terminal-Bench 2.0
GLM-5.281.0%SourceKimi K2.666.7%SourceGLM-5.2 leads this result
ProgramBench
GLM-5.263.7%SourceKimi K2.6—Not directly comparable
cursorBench32
GLM-5.255.0%SourceKimi K2.6—Not directly comparable
OpenHarmony Bench
GLM-5.258.4%SourceKimi K2.6—Not directly comparable
LiveCodeBench (Vals)
GLM-5.269.5%SourceKimi K2.686.8%SourceKimi K2.6 leads this result
SWE-bench (Vals)
GLM-5.282.8%SourceKimi K2.676.2%SourceGLM-5.2 leads this result
SWE-bench Verified
GLM-5.2—Kimi K2.680.2%SourceNot directly comparable
LiveCodeBench v6
GLM-5.2—Kimi K2.689.6%SourceNot directly comparable
SWE Multilingual
GLM-5.2—Kimi K2.676.7%SourceNot directly comparable
SciCode
GLM-5.2—Kimi K2.652.2%SourceNot directly comparable
Vibe Code Bench
GLM-5.2—Kimi K2.637.89%SourceNot directly comparable
cursorBench31
GLM-5.2—Kimi K2.647.6%SourceNot directly comparable
Reasoning
CritPt
GLM-5.220.9%SourceKimi K2.6—Not directly comparable
Knowledge
GPQA
GLM-5.291.2%SourceKimi K2.690.5%SourceGLM-5.2 leads this result
GPQA-D
GLM-5.291.2%SourceKimi K2.690.5%SourceGLM-5.2 leads this result
HLE
GLM-5.254.7%SourceKimi K2.634.7%SourceGLM-5.2 leads this result
HLE w/o tools
GLM-5.240.5%SourceKimi K2.6—Not directly comparable
GPQA Diamond (Vals)
GLM-5.285.6%SourceKimi K2.689.1%SourceKimi K2.6 leads this result
MMLU-Pro (Vals)
GLM-5.286.7%SourceKimi K2.687.6%SourceKimi K2.6 leads this result
Math
AIME26
GLM-5.299.2%SourceKimi K2.696.4%SourceGLM-5.2 leads this result
HMMT Nov 2025
GLM-5.294.4%SourceKimi K2.6—Not directly comparable
HMMT Feb 2026
GLM-5.292.5%SourceKimi K2.692.7%SourceKimi K2.6 leads this result
MMAnswerBench
GLM-5.291.0%SourceKimi K2.686.0%SourceGLM-5.2 leads this result
FrontierMath v2 (Tiers 1-3)
GLM-5.2—Kimi K2.638.966%SourceNot directly comparable
FrontierMath v2 (Tier 4)
GLM-5.2—Kimi K2.614.580%SourceNot directly comparable
Multimodal
MMMU-Pro
GLM-5.2—Kimi K2.679.4%SourceNot directly comparable
MMMU-Pro w/ Python
GLM-5.2—Kimi K2.680.1%SourceNot directly comparable
CharXiv
GLM-5.2—Kimi K2.680.4%SourceNot directly comparable
MathVision
GLM-5.2—Kimi K2.687.4%SourceNot directly comparable
V*
GLM-5.2—Kimi K2.696.9%SourceNot directly comparable
Frequently asked questions
Which is better, GLM-5.2 or Kimi K2.6?
GLM-5.2 has the higher public score estimate, 68.19 versus 65.46, 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-5.2 or Kimi K2.6?
GLM-5.2 leads the public coding lane, 61 to 50.8, with Supported evidence for both models, although the 90% intervals overlap.
Which is better for agentic tasks, GLM-5.2 or Kimi K2.6?
GLM-5.2 leads the public agentic tasks lane, 58.5 to 46.6, with Supported evidence for both models, although the 90% intervals overlap.
Which costs less, GLM-5.2 or Kimi K2.6?
For the stated presets, chat costs $0.0036 on GLM-5.2 and $0.00295 on Kimi K2.6; repository review costs $0.0832 and $0.0595; the cache-heavy agent loop costs $0.352 and $0.249. GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate. Kimi K2.6 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, GLM-5.2 or Kimi K2.6?
GLM-5.2 has the larger documented context window: 1M, compared with 256K.
Related comparisons
Last updated September 14, 2026
Watch GLM-5.2 vs Kimi K2.6
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.