Agentic
Not comparable- Claude 3.5 Sonnet
- Not ranked
- Kimi K3
- 71.9
- Supported · #4/152
- Basis
- BenchAlign lane · 0 vs 11 public rows
- Reading
- Not comparable
Claude 3.5 Sonnet vs Kimi K3
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Kimi K3 has the higher public score, 74.76 versus 32.01, and the 90% score intervals do not overlap.
1 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.
Long documents
Prompts that approach the documented context limit
Kimi K3
Kimi K3 has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Claude 3.5 Sonnet 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
Claude 3.5 Sonnet is not ranked on the public lane for agentic, so no winner is named for agentic.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
No clear pick
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
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. Claude 3.5 Sonnet does not fit this workload in one request. Claude 3.5 Sonnet 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
No clear pick
A complete comparable API-rate estimate is not available for both models.
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.
3 Claude 3.5 Sonnet only
1 Shared
42 Kimi K3 only
- Shared results
- 1
- Claude 3.5 Sonnet only
- 3
- Kimi K3 only
- 42
- 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- Claude 3.5 Sonnet
- Not ranked
- Kimi K3
- 68.0
- Supported · #6/151
- Basis
- BenchAlign lane · 1 vs 13 public rows
- Reading
- Not comparable
Reasoning
Not comparable- Claude 3.5 Sonnet
- Not ranked
- Kimi K3
- 78.5
- #3/18
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Claude 3.5 Sonnet
- Not ranked
- Kimi K3
- 72.1
- Supported · #8/182
- Basis
- BenchAlign lane · 1 vs 6 public rows
- Reading
- Not comparable
Math
Not comparable- Claude 3.5 Sonnet
- 25.7
- Unranked · 2 rankable rows
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Claude 3.5 Sonnet
- Not ranked
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Claude 3.5 Sonnet
- Not ranked
- Kimi K3
- 89.5
- #1/48
- Basis
- Provisional lane · 0 vs 3 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Claude 3.5 Sonnet
- Not ranked
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Claude 3.5 Sonnet | Kimi K3 | Basis | Reading |
|---|---|---|---|---|
| Agentic | Not ranked | 71.9Supported · #4/152 | Not comparableBenchAlign lane · 0 vs 11 public rows | Not comparable |
| Coding | Not ranked | 68.0Supported · #6/151 | Not comparableBenchAlign lane · 1 vs 13 public rows | Not comparable |
| Reasoning | Not ranked | 78.5#3/18 | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Knowledge | Not ranked | 72.1Supported · #8/182 | Not comparableBenchAlign lane · 1 vs 6 public rows | Not comparable |
| Math | 25.7Unranked · 2 rankable rows | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | Not ranked | 89.5#1/48 | Not comparableProvisional lane · 0 vs 3 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.
GPQA
Knowledge
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
- Claude 3.5 Sonnet
- $0.0105
- Fits in one request
- Kimi K3
- $0.0105
- Fits in one request
Modeled costs are equal
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Claude 3.5 Sonnet
- $0.195
- Fits in one request
- Kimi K3
- $0.195
- Fits in one request
Modeled costs are equal
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Claude 3.5 Sonnet
- $0.81
- Does not fit in one request
- Cached input priced at the published list-input rate
- Kimi K3
- $0.27
- Fits in one request
Claude 3.5 Sonnet does not fit this workload in one request. Claude 3.5 Sonnet 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.
SpecificationClaude 3.5 SonnetKimi K3
Context window
Maximum documented context; output-token limits may be lower.
Claude 3.5 Sonnet
200K
Kimi K3
1.05M
API model ID
Claude 3.5 Sonnet
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.
Claude 3.5 Sonnet
Not published
Kimi K3
$0.3 per 1M cached input tokens
Documented inputs
Claude 3.5 Sonnet
Not sourced
Kimi K3
Not sourced
Documented outputs
Claude 3.5 Sonnet
Not sourced
Kimi K3
Not sourced
Provider availability
Claude 3.5 Sonnet
Not sourced
Kimi K3
Not sourced
Reasoning profile
Claude 3.5 Sonnet
Non-Reasoning
Kimi K3
Reasoning
Weight access
Claude 3.5 Sonnet
Proprietary
Kimi K3
Pending
License
Claude 3.5 Sonnet
Proprietary
Kimi K3
Pending
Release date
Claude 3.5 Sonnet
2024-06-01
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, 74.76 versus 32.01, and the 90% score intervals do not overlap.
- Workload cost
- Repository review: $0.195 vs $0.195. Cache-heavy agent loop: $0.81 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.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence46 rows
Agentic
Terminal-Bench 2.0
Claude 3.5 Sonnet—Kimi K388.3%SourceNot directly comparable
BrowseComp
Claude 3.5 Sonnet—Kimi K391.2%SourceNot directly comparable
DeepSearchQA
Claude 3.5 Sonnet—Kimi K395.0%SourceNot directly comparable
Toolathlon-Verified
Claude 3.5 Sonnet—Kimi K373.2%SourceNot directly comparable
MCP Atlas
Claude 3.5 Sonnet—Kimi K384.2%SourceNot directly comparable
AutomationBench
Claude 3.5 Sonnet—Kimi K330.8%SourceNot directly comparable
JobBench
Claude 3.5 Sonnet—Kimi K352.9%SourceNot directly comparable
APEX-Agents
Claude 3.5 Sonnet—Kimi K337.6%SourceNot directly comparable
SpreadsheetBench 2
Claude 3.5 Sonnet—Kimi K334.8%SourceNot directly comparable
DECK-Bench
Claude 3.5 Sonnet—Kimi K373.5%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Claude 3.5 Sonnet—Kimi K380.9%SourceNot directly comparable
Coding
SWE-bench Verified
Claude 3.5 Sonnet49%SourceKimi K3—Not directly comparable
DeepSWE
Claude 3.5 Sonnet—Kimi K367.5%SourceNot directly comparable
cursorBench32
Claude 3.5 Sonnet—Kimi K360.8%SourceNot directly comparable
FrontierSWE
Claude 3.5 Sonnet—Kimi K381.2%SourceNot directly comparable
ProgramBench
Claude 3.5 Sonnet—Kimi K377.8%SourceNot directly comparable
Kimi Code Bench v2
Claude 3.5 Sonnet—Kimi K372.9%SourceNot directly comparable
sweMarathon
Claude 3.5 Sonnet—Kimi K342%SourceNot directly comparable
PostTrain Bench
Claude 3.5 Sonnet—Kimi K336.6%SourceNot directly comparable
MLS-Bench Lite
Claude 3.5 Sonnet—Kimi K348.3%SourceNot directly comparable
VulcanBench v3
Claude 3.5 Sonnet—Kimi K373.7%SourceNot directly comparable
OpenHarmony Bench
Claude 3.5 Sonnet—Kimi K357.3%SourceNot directly comparable
FrontierSWE v2
Claude 3.5 Sonnet—Kimi K325.9%SourceNot directly comparable
LiveCodeBench (Vals)
Claude 3.5 Sonnet—Kimi K387.2%SourceNot directly comparable
SWE-bench (Vals)
Claude 3.5 Sonnet—Kimi K393.4%SourceNot directly comparable
Knowledge
GPQA
Claude 3.5 Sonnet59.4%SourceKimi K393.5%SourceKimi K3 leads this result
GPQA-D
Claude 3.5 Sonnet—Kimi K393.5%SourceNot directly comparable
HLE
Claude 3.5 Sonnet—Kimi K356%SourceNot directly comparable
HLE w/o tools
Claude 3.5 Sonnet—Kimi K343.5%SourceNot directly comparable
GPQA Diamond (Vals)
Claude 3.5 Sonnet—Kimi K392.9%SourceNot directly comparable
MMLU-Pro (Vals)
Claude 3.5 Sonnet—Kimi K388.0%SourceNot directly comparable
Math
Multimodal
OfficeQA Pro
Claude 3.5 Sonnet—Kimi K363.3%SourceNot directly comparable
MMMU-Pro
Claude 3.5 Sonnet—Kimi K381.6%SourceNot directly comparable
MMMU-Pro w/ Python
Claude 3.5 Sonnet—Kimi K383.4%SourceNot directly comparable
CharXiv w/o tools
Claude 3.5 Sonnet—Kimi K384.8%SourceNot directly comparable
CharXiv
Claude 3.5 Sonnet—Kimi K391.3%SourceNot directly comparable
MathVision
Claude 3.5 Sonnet—Kimi K394.3%SourceNot directly comparable
MathVision w/ Python
Claude 3.5 Sonnet—Kimi K397.8%SourceNot directly comparable
BabyVision w/ Python
Claude 3.5 Sonnet—Kimi K385.7%SourceNot directly comparable
ZeroBench
Claude 3.5 Sonnet—Kimi K323.0%SourceNot directly comparable
ZeroBench w/ Python
Claude 3.5 Sonnet—Kimi K341.0%SourceNot directly comparable
WorldVQA ForceAnswer
Claude 3.5 Sonnet—Kimi K351.0%SourceNot directly comparable
OmniDocBench
Claude 3.5 Sonnet—Kimi K391.1%SourceNot directly comparable
PerceptionBench
Claude 3.5 Sonnet—Kimi K358.5%SourceNot directly comparable
Frequently asked questions
Which is better, Claude 3.5 Sonnet or Kimi K3?
Kimi K3 has the higher public score, 74.76 versus 32.01, 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, Claude 3.5 Sonnet or Kimi K3?
Claude 3.5 Sonnet is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, Claude 3.5 Sonnet or Kimi K3?
Claude 3.5 Sonnet is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, Claude 3.5 Sonnet or Kimi K3?
For the stated presets, chat costs $0.0105 on Claude 3.5 Sonnet and $0.0105 on Kimi K3; repository review costs $0.195 and $0.195; the cache-heavy agent loop costs $0.81 and $0.27. Claude 3.5 Sonnet does not fit this workload in one request. Claude 3.5 Sonnet has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Claude 3.5 Sonnet or Kimi K3?
Kimi K3 has the larger documented context window: 1.05M, compared with 200K.
Related comparisons
Last updated September 10, 2026
Watch Claude 3.5 Sonnet 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.