Agentic
Not comparable- Kimi K3
- 72.0
- Supported · #4/154
- Pareto 26.9
- Not ranked
- Basis
- BenchAlign lane · 12 vs 1 public rows
- Reading
- Not comparable
Kimi K3 vs Pareto 26.9
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
3 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Updated September 18, 2026. We do not rank this pair: at least one has no public score. 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
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
Pareto 26.9
Pareto 26.9 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
Pareto 26.9
Pareto 26.9 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
Pareto 26.9
Pareto 26.9 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
Pareto 26.9 is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Pareto 26.9 is not ranked on the public lane for agentic, so no winner is named for agentic.
Confidence: limited
Long documents
Prompts that approach the documented context limit
Not enough matched evidence
A complete context comparison is not sourced.
Confidence: limited
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
41 Kimi K3 only
3 Shared
1 Pareto 26.9 only
- Shared results
- 3
- Kimi K3 only
- 41
- Pareto 26.9 only
- 1
- Like-for-like categories
- 0 / 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
Not comparable- Kimi K3
- 67.7
- Supported · #6/154
- Pareto 26.9
- Not ranked
- Basis
- BenchAlign lane · 13 vs 1 public rows
- Reading
- Not comparable
Reasoning
Not comparable- Kimi K3
- 78.5
- #3/20
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Kimi K3
- 71.9
- Supported · #8/184
- Pareto 26.9
- Not ranked
- Basis
- BenchAlign lane · 6 vs 1 public rows
- Reading
- Not comparable
Math
Not comparable- Kimi K3
- Not ranked
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K3
- Not ranked
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Kimi K3
- 89.4
- #1/48
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 3 vs 1 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Kimi K3
- Not ranked
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Kimi K3 | Pareto 26.9 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 72.0Supported · #4/154 | Not ranked | Not comparableBenchAlign lane · 12 vs 1 public rows | Not comparable |
| Coding | 67.7Supported · #6/154 | Not ranked | Not comparableBenchAlign lane · 13 vs 1 public rows | Not comparable |
| Reasoning | 78.5#3/20 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Knowledge | 71.9Supported · #8/184 | Not ranked | Not comparableBenchAlign lane · 6 vs 1 public rows | Not comparable |
| Math | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | 89.4#1/48 | Not ranked | Not comparableProvisional lane · 3 vs 1 weighted rows | Not comparable |
| Instruction following | Not ranked | 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 w/o tools
Knowledge
MMMU-Pro
Multimodal
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
- Kimi K3
- $0.0105
- Fits in one request
- Pareto 26.9
- $0.00625
- Fit state unavailable
Pareto 26.9 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Kimi K3
- $0.195
- Fits in one request
- Pareto 26.9
- $0.1475
- Fit state unavailable
Pareto 26.9 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Kimi K3
- $0.27
- Fits in one request
- Pareto 26.9
- $0.175
- Fit state unavailable
Pareto 26.9 has the lower modeled cost
Costs use the listed standard API rates.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationKimi K3Pareto 26.9
Context window
Maximum documented context; output-token limits may be lower.
Kimi K3
1.05M
Pareto 26.9
N/A
API model ID
Kimi K3
Not sourced
Pareto 26.9
pareto
Unbiased pricingCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Kimi K3
$0.3 per 1M cached input tokens
Pareto 26.9
$0.25 per 1M cached input tokens
Unbiased pricingDocumented inputs
Kimi K3
Not sourced
Pareto 26.9
Not sourced
Documented outputs
Kimi K3
Not sourced
Pareto 26.9
Not sourced
Provider availability
Kimi K3
Not sourced
Pareto 26.9
Not sourced
Reasoning profile
Kimi K3
Reasoning
Pareto 26.9
Reasoning
Weight access
Kimi K3
Pending
Pareto 26.9
Proprietary
License
Kimi K3
Pending
Pareto 26.9
Proprietary
Release date
Kimi K3
2026-07-16
Pareto 26.9
2026-09-17
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
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- Workload cost
- Repository review: $0.195 vs $0.1475. Cache-heavy agent loop: $0.27 vs $0.175.
- Context tradeoff
- A complete documented context comparison is not available.
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 evidence45 rows
Agentic
Terminal-Bench 2.0
Kimi K388.3%SourcePareto 26.9—Not directly comparable
BrowseComp
Kimi K391.2%SourcePareto 26.9—Not directly comparable
DeepSearchQA
Kimi K395.0%SourcePareto 26.9—Not directly comparable
Toolathlon-Verified
Kimi K373.2%SourcePareto 26.9—Not directly comparable
MCP Atlas
Kimi K384.2%SourcePareto 26.9—Not directly comparable
AutomationBench
Kimi K330.8%SourcePareto 26.9—Not directly comparable
JobBench
Kimi K352.9%SourcePareto 26.9—Not directly comparable
APEX-Agents
Kimi K337.6%SourcePareto 26.9—Not directly comparable
SpreadsheetBench 2
Kimi K334.8%SourcePareto 26.9—Not directly comparable
DECK-Bench
Kimi K373.5%SourcePareto 26.9—Not directly comparable
Terminal-Bench 2.1 (Vals)
Kimi K380.9%SourcePareto 26.9—Not directly comparable
ApprenticeBench
Kimi K318%SourcePareto 26.9—Not directly comparable
Terminal-Bench 4.0
Kimi K3—Pareto 26.951.00%SourceNot directly comparable
Coding
DeepSWE
Kimi K367.5%SourcePareto 26.974.0%SourcePareto 26.9 leads this result
cursorBench32
Kimi K360.8%SourcePareto 26.9—Not directly comparable
FrontierSWE
Kimi K381.2%SourcePareto 26.9—Not directly comparable
ProgramBench
Kimi K377.8%SourcePareto 26.9—Not directly comparable
Kimi Code Bench v2
Kimi K372.9%SourcePareto 26.9—Not directly comparable
sweMarathon
Kimi K342%SourcePareto 26.9—Not directly comparable
PostTrain Bench
Kimi K336.6%SourcePareto 26.9—Not directly comparable
MLS-Bench Lite
Kimi K348.3%SourcePareto 26.9—Not directly comparable
VulcanBench v3
Kimi K373.7%SourcePareto 26.9—Not directly comparable
OpenHarmony Bench
Kimi K357.3%SourcePareto 26.9—Not directly comparable
FrontierSWE v2
Kimi K325.9%SourcePareto 26.9—Not directly comparable
LiveCodeBench (Vals)
Kimi K387.2%SourcePareto 26.9—Not directly comparable
SWE-bench (Vals)
Kimi K393.4%SourcePareto 26.9—Not directly comparable
Knowledge
GPQA
Kimi K393.5%SourcePareto 26.9—Not directly comparable
GPQA-D
Kimi K393.5%SourcePareto 26.9—Not directly comparable
HLE
Kimi K356%SourcePareto 26.9—Not directly comparable
HLE w/o tools
Kimi K343.5%SourcePareto 26.949%SourcePareto 26.9 leads this result
GPQA Diamond (Vals)
Kimi K392.9%SourcePareto 26.9—Not directly comparable
MMLU-Pro (Vals)
Kimi K388.0%SourcePareto 26.9—Not directly comparable
Multimodal
OfficeQA Pro
Kimi K363.3%SourcePareto 26.9—Not directly comparable
MMMU-Pro
Kimi K381.6%SourcePareto 26.978%SourceKimi K3 leads this result
MMMU-Pro w/ Python
Kimi K383.4%SourcePareto 26.9—Not directly comparable
CharXiv w/o tools
Kimi K384.8%SourcePareto 26.9—Not directly comparable
CharXiv
Kimi K391.3%SourcePareto 26.9—Not directly comparable
MathVision
Kimi K394.3%SourcePareto 26.9—Not directly comparable
MathVision w/ Python
Kimi K397.8%SourcePareto 26.9—Not directly comparable
BabyVision w/ Python
Kimi K385.7%SourcePareto 26.9—Not directly comparable
ZeroBench
Kimi K323.0%SourcePareto 26.9—Not directly comparable
ZeroBench w/ Python
Kimi K341.0%SourcePareto 26.9—Not directly comparable
WorldVQA ForceAnswer
Kimi K351.0%SourcePareto 26.9—Not directly comparable
OmniDocBench
Kimi K391.1%SourcePareto 26.9—Not directly comparable
PerceptionBench
Kimi K358.5%SourcePareto 26.9—Not directly comparable
Questions
Which is better, Kimi K3 or Pareto 26.9?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, Kimi K3 or Pareto 26.9?
Pareto 26.9 is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, Kimi K3 or Pareto 26.9?
Pareto 26.9 is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, Kimi K3 or Pareto 26.9?
For the stated presets, chat costs $0.0105 on Kimi K3 and $0.00625 on Pareto 26.9; repository review costs $0.195 and $0.1475; the cache-heavy agent loop costs $0.27 and $0.175. Costs use the listed standard API rates.
Which has the larger context window, Kimi K3 or Pareto 26.9?
A complete documented context-window comparison is not available.
Related comparisons
Last updated September 18, 2026
Watch Kimi K3 vs Pareto 26.9
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.