Agentic
Not comparable- Agents-A1-4B
- Not ranked
- Kimi K3
- 71.9
- Supported · #4/152
- Basis
- BenchAlign lane · 5 vs 11 public rows
- Reading
- Not comparable
Agents-A1-4B vs Kimi K3
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
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.
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
Agents-A1-4B 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
Agents-A1-4B 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
Not enough matched evidence
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
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
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.
10 Agents-A1-4B only
1 Shared
42 Kimi K3 only
- Shared results
- 1
- Agents-A1-4B only
- 10
- 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- Agents-A1-4B
- Not ranked
- Kimi K3
- 68.0
- Supported · #6/151
- Basis
- BenchAlign lane · 2 vs 13 public rows
- Reading
- Not comparable
Reasoning
Not comparable- Agents-A1-4B
- Not ranked
- Kimi K3
- 78.5
- #3/20
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Agents-A1-4B
- Not ranked
- Kimi K3
- 72.0
- Supported · #8/183
- Basis
- BenchAlign lane · 1 vs 6 public rows
- Reading
- Not comparable
Math
Not comparable- Agents-A1-4B
- Not ranked
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Agents-A1-4B
- Not ranked
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Agents-A1-4B
- Not ranked
- Kimi K3
- 89.5
- #1/48
- Basis
- Provisional lane · 0 vs 3 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Agents-A1-4B
- Not ranked
- Kimi K3
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
| Category | Agents-A1-4B | Kimi K3 | Basis | Reading |
|---|---|---|---|---|
| Agentic | Not ranked | 71.9Supported · #4/152 | Not comparableBenchAlign lane · 5 vs 11 public rows | Not comparable |
| Coding | Not ranked | 68.0Supported · #6/151 | Not comparableBenchAlign lane · 2 vs 13 public rows | Not comparable |
| Reasoning | Not ranked | 78.5#3/20 | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Knowledge | Not ranked | 72.0Supported · #8/183 | Not comparableBenchAlign lane · 1 vs 6 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 | 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 · 1 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.
BrowseComp
Agentic
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
- Agents-A1-4B
- Self-hosted; infrastructure cost varies
- Fits in one request
- Kimi K3
- $0.0105
- Fits in one request
Agents-A1-4B has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Agents-A1-4B
- Self-hosted; infrastructure cost varies
- Fits in one request
- Kimi K3
- $0.195
- Fits in one request
Agents-A1-4B has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Agents-A1-4B
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
- Kimi K3
- $0.27
- Fits in one request
Agents-A1-4B has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationAgents-A1-4BKimi K3
Context window
Maximum documented context; output-token limits may be lower.
Agents-A1-4B
Kimi K3
1.05M
API model ID
Agents-A1-4B
InternScience/Agents-A1-4B
InternScience Agents-A1-4B model cardKimi K3
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Agents-A1-4B
No comparable hosted API rate
InternScience Agents-A1-4B model cardKimi K3
$0.3 per 1M cached input tokens
Documented inputs
Agents-A1-4B
Not sourced
Kimi K3
Not sourced
Documented outputs
Agents-A1-4B
Not sourced
Kimi K3
Not sourced
Provider availability
Agents-A1-4B
Not sourced
Kimi K3
Not sourced
Reasoning profile
Agents-A1-4B
Reasoning
Kimi K3
Reasoning
Weight access
Agents-A1-4B
Open Weight
Kimi K3
Pending
License
Agents-A1-4B
Open Weight
Kimi K3
Pending
Release date
Agents-A1-4B
2026-07-13
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
- 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
- A complete comparable API-rate estimate is not available for both models.
- 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 evidence53 rows
Agentic
BrowseComp
Agents-A1-4B66.8%SourceKimi K391.2%SourceKimi K3 leads this result
GAIA
Agents-A1-4B95.1%SourceKimi K3—Not directly comparable
MLE-Bench Lite
Agents-A1-4B22.7%SourceKimi K3—Not directly comparable
τ²-bench results
Agents-A1-4B78.2%SourceKimi K3—Not directly comparable
VITA-Bench
Agents-A1-4B40.3%SourceKimi K3—Not directly comparable
Terminal-Bench 2.0
Agents-A1-4B—Kimi K388.3%SourceNot directly comparable
DeepSearchQA
Agents-A1-4B—Kimi K395.0%SourceNot directly comparable
Toolathlon-Verified
Agents-A1-4B—Kimi K373.2%SourceNot directly comparable
MCP Atlas
Agents-A1-4B—Kimi K384.2%SourceNot directly comparable
AutomationBench
Agents-A1-4B—Kimi K330.8%SourceNot directly comparable
JobBench
Agents-A1-4B—Kimi K352.9%SourceNot directly comparable
APEX-Agents
Agents-A1-4B—Kimi K337.6%SourceNot directly comparable
SpreadsheetBench 2
Agents-A1-4B—Kimi K334.8%SourceNot directly comparable
DECK-Bench
Agents-A1-4B—Kimi K373.5%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Agents-A1-4B—Kimi K380.9%SourceNot directly comparable
Coding
SciCode
Agents-A1-4B29.6%SourceKimi K3—Not directly comparable
LiveCodeBench v6
Agents-A1-4B59.6%SourceKimi K3—Not directly comparable
DeepSWE
Agents-A1-4B—Kimi K367.5%SourceNot directly comparable
cursorBench32
Agents-A1-4B—Kimi K360.8%SourceNot directly comparable
FrontierSWE
Agents-A1-4B—Kimi K381.2%SourceNot directly comparable
ProgramBench
Agents-A1-4B—Kimi K377.8%SourceNot directly comparable
Kimi Code Bench v2
Agents-A1-4B—Kimi K372.9%SourceNot directly comparable
sweMarathon
Agents-A1-4B—Kimi K342%SourceNot directly comparable
PostTrain Bench
Agents-A1-4B—Kimi K336.6%SourceNot directly comparable
MLS-Bench Lite
Agents-A1-4B—Kimi K348.3%SourceNot directly comparable
VulcanBench v3
Agents-A1-4B—Kimi K373.7%SourceNot directly comparable
OpenHarmony Bench
Agents-A1-4B—Kimi K357.3%SourceNot directly comparable
FrontierSWE v2
Agents-A1-4B—Kimi K325.9%SourceNot directly comparable
LiveCodeBench (Vals)
Agents-A1-4B—Kimi K387.2%SourceNot directly comparable
SWE-bench (Vals)
Agents-A1-4B—Kimi K393.4%SourceNot directly comparable
Reasoning
LongBench v2
Agents-A1-4B52.1%SourceKimi K3—Not directly comparable
Knowledge
FrontierScience Research
Agents-A1-4B33.3%SourceKimi K3—Not directly comparable
GPQA
Agents-A1-4B—Kimi K393.5%SourceNot directly comparable
GPQA-D
Agents-A1-4B—Kimi K393.5%SourceNot directly comparable
HLE
Agents-A1-4B—Kimi K356%SourceNot directly comparable
HLE w/o tools
Agents-A1-4B—Kimi K343.5%SourceNot directly comparable
GPQA Diamond (Vals)
Agents-A1-4B—Kimi K392.9%SourceNot directly comparable
MMLU-Pro (Vals)
Agents-A1-4B—Kimi K388.0%SourceNot directly comparable
Multimodal
OfficeQA Pro
Agents-A1-4B—Kimi K363.3%SourceNot directly comparable
MMMU-Pro
Agents-A1-4B—Kimi K381.6%SourceNot directly comparable
MMMU-Pro w/ Python
Agents-A1-4B—Kimi K383.4%SourceNot directly comparable
CharXiv w/o tools
Agents-A1-4B—Kimi K384.8%SourceNot directly comparable
CharXiv
Agents-A1-4B—Kimi K391.3%SourceNot directly comparable
MathVision
Agents-A1-4B—Kimi K394.3%SourceNot directly comparable
MathVision w/ Python
Agents-A1-4B—Kimi K397.8%SourceNot directly comparable
BabyVision w/ Python
Agents-A1-4B—Kimi K385.7%SourceNot directly comparable
ZeroBench
Agents-A1-4B—Kimi K323.0%SourceNot directly comparable
ZeroBench w/ Python
Agents-A1-4B—Kimi K341.0%SourceNot directly comparable
WorldVQA ForceAnswer
Agents-A1-4B—Kimi K351.0%SourceNot directly comparable
OmniDocBench
Agents-A1-4B—Kimi K391.1%SourceNot directly comparable
PerceptionBench
Agents-A1-4B—Kimi K358.5%SourceNot directly comparable
Instruction following
Frequently asked questions
Which is better, Agents-A1-4B or Kimi K3?
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, Agents-A1-4B or Kimi K3?
Agents-A1-4B is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, Agents-A1-4B or Kimi K3?
Agents-A1-4B is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, Agents-A1-4B or Kimi K3?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, Agents-A1-4B or Kimi K3?
Kimi K3 has the larger documented context window: 1.05M, compared with 262K.
Related comparisons
Last updated September 10, 2026
Watch Agents-A1-4B 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.