Agentic
Directional only- Kimi K3
- 89.5
- MiMo-V2.5
- 65.8
- Weighted basis
- 2 vs 1 rows
- Reading
- Directional only
Kimi K3 vs MiMo-V2.5
Updated August 19, 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, 80.53 versus 59.15, and the 90% score intervals do not overlap.
3 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.
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
No shared weighted benchmark basis supports a winner.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
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.
37 Kimi K3 only
3 Shared
7 MiMo-V2.5 only
- Shared results
- 3
- Kimi K3 only
- 37
- MiMo-V2.5 only
- 7
- Like-for-like categories
- 0 / 8
2 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.
Multimodal
Directional only- Kimi K3
- 78.5
- MiMo-V2.5
- 79.0
- Weighted basis
- 3 vs 2 rows
- Reading
- Directional only
Coding
Not comparable- Kimi K3
- Not measured
- MiMo-V2.5
- 56.1
- Weighted basis
- 0 vs 1 rows
- Reading
- Not comparable
Reasoning
Not comparable- Kimi K3
- Not measured
- MiMo-V2.5
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Knowledge
Not comparable- Kimi K3
- 61.0
- MiMo-V2.5
- Not measured
- Weighted basis
- 2 vs 0 rows
- Reading
- Not comparable
Math
Not comparable- Kimi K3
- Not measured
- MiMo-V2.5
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K3
- Not measured
- MiMo-V2.5
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Instruction following
Not comparable- Kimi K3
- Not measured
- MiMo-V2.5
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | Kimi K3 | MiMo-V2.5 | Weighted basis | Reading |
|---|---|---|---|---|
| Agentic | 89.5 | 65.8 | Directional only2 vs 1 rows | Directional only |
| Multimodal | 78.5 | 79.0 | Directional only3 vs 2 rows | Directional only |
| Coding | Not measured | 56.1 | Not comparable0 vs 1 rows | Not comparable |
| Reasoning | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Knowledge | 61.0 | Not measured | Not comparable2 vs 0 rows | Not comparable |
| Math | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multilingual | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Instruction following | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
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.
Terminal-Bench 2.0
Agentic
CharXiv
Multimodal
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
- MiMo-V2.5
- API rate not published
- Fits in one request
MiMo-V2.5 has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Kimi K3
- $0.195
- Fits in one request
- MiMo-V2.5
- API rate not published
- Fits in one request
MiMo-V2.5 has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Kimi K3
- $0.27
- Fits in one request
- MiMo-V2.5
- API rate not published
- Fits in one request
- Cached-input rate unavailable
MiMo-V2.5 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.
SpecificationKimi K3MiMo-V2.5
Context window
Maximum documented context; output-token limits may be lower.
Kimi K3
1.05M
MiMo-V2.5
1M
API model ID
Kimi K3
Not sourced
MiMo-V2.5
Not sourced
Cached-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
MiMo-V2.5
No comparable hosted API rate
Documented inputs
Kimi K3
Not sourced
MiMo-V2.5
Not sourced
Documented outputs
Kimi K3
Not sourced
MiMo-V2.5
Not sourced
Provider availability
Kimi K3
Not sourced
MiMo-V2.5
Not sourced
Reasoning profile
Kimi K3
Reasoning
MiMo-V2.5
Reasoning
Weight access
Kimi K3
Pending
MiMo-V2.5
Proprietary
License
Kimi K3
Pending
MiMo-V2.5
Proprietary
Release date
Kimi K3
2026-07-16
MiMo-V2.5
2026-04-22
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, 80.53 versus 59.15, and the 90% score intervals do not overlap.
- 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 evidence47 rows
Agentic
Terminal-Bench 2.0
Kimi K388.3%SourceMiMo-V2.565.8%SourceKimi K3 leads this result
BrowseComp
Kimi K391.2%SourceMiMo-V2.5—Not directly comparable
DeepSearchQA
Kimi K395.0%SourceMiMo-V2.5—Not directly comparable
Toolathlon-Verified
Kimi K373.2%SourceMiMo-V2.5—Not directly comparable
MCP Atlas
Kimi K384.2%SourceMiMo-V2.5—Not directly comparable
AutomationBench
Kimi K330.8%SourceMiMo-V2.5—Not directly comparable
JobBench
Kimi K352.9%SourceMiMo-V2.5—Not directly comparable
APEX-Agents
Kimi K337.6%SourceMiMo-V2.5—Not directly comparable
SpreadsheetBench 2
Kimi K334.8%SourceMiMo-V2.5—Not directly comparable
DECK-Bench
Kimi K373.5%SourceMiMo-V2.5—Not directly comparable
Claw-Eval
Kimi K3—MiMo-V2.562.3%SourceNot directly comparable
MM-ClawBench
Kimi K3—MiMo-V2.523.8%SourceNot directly comparable
Gert Labs
Kimi K3—MiMo-V2.546.89%SourceNot directly comparable
ResearchClawBench
Kimi K3—MiMo-V2.516.9%SourceNot directly comparable
Coding
deepSwe
Kimi K367.5%SourceMiMo-V2.5—Not directly comparable
cursorBench32
Kimi K360.8%SourceMiMo-V2.5—Not directly comparable
FrontierSWE
Kimi K381.2%SourceMiMo-V2.5—Not directly comparable
ProgramBench
Kimi K377.8%SourceMiMo-V2.5—Not directly comparable
Kimi Code Bench v2
Kimi K372.9%SourceMiMo-V2.5—Not directly comparable
sweMarathon
Kimi K342%SourceMiMo-V2.5—Not directly comparable
PostTrain Bench
Kimi K336.6%SourceMiMo-V2.5—Not directly comparable
MLS-Bench Lite
Kimi K348.3%SourceMiMo-V2.5—Not directly comparable
VulcanBench v3
Kimi K373.9%SourceMiMo-V2.5—Not directly comparable
APEX-SWE
Kimi K348.0%SourceMiMo-V2.5—Not directly comparable
EEBench
Kimi K338.3%SourceMiMo-V2.5—Not directly comparable
InferenceEval
Kimi K339.8%SourceMiMo-V2.5—Not directly comparable
KernelBench Internal
Kimi K368.5%SourceMiMo-V2.5—Not directly comparable
SWE-bench Pro
Kimi K3—MiMo-V2.556.1%SourceNot directly comparable
Terminal-Bench 2.0
Kimi K3—MiMo-V2.565.8%SourceNot directly comparable
Knowledge
Multimodal
OfficeQA Pro
Kimi K363.3%SourceMiMo-V2.5—Not directly comparable
MMMU-Pro
Kimi K381.6%SourceMiMo-V2.577.9%SourceKimi K3 leads this result
MMMU-Pro w/ Python
Kimi K383.4%SourceMiMo-V2.5—Not directly comparable
CharXiv w/o tools
Kimi K384.8%SourceMiMo-V2.5—Not directly comparable
CharXiv
Kimi K391.3%SourceMiMo-V2.581%SourceKimi K3 leads this result
MathVision
Kimi K394.3%SourceMiMo-V2.5—Not directly comparable
MathVision w/ Python
Kimi K397.8%SourceMiMo-V2.5—Not directly comparable
BabyVision w/ Python
Kimi K385.7%SourceMiMo-V2.5—Not directly comparable
ZeroBench
Kimi K323.0%SourceMiMo-V2.5—Not directly comparable
ZeroBench w/ Python
Kimi K341.0%SourceMiMo-V2.5—Not directly comparable
WorldVQA ForceAnswer
Kimi K351.0%SourceMiMo-V2.5—Not directly comparable
OmniDocBench
Kimi K391.1%SourceMiMo-V2.5—Not directly comparable
PerceptionBench
Kimi K358.5%SourceMiMo-V2.5—Not directly comparable
Video-MME (with subtitle)
Kimi K3—MiMo-V2.587.7%SourceNot directly comparable
Frequently asked questions
Which is better, Kimi K3 or MiMo-V2.5?
Kimi K3 has the higher public score, 80.53 versus 59.15, 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, Kimi K3 or MiMo-V2.5?
The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.
Which is better for agentic tasks, Kimi K3 or MiMo-V2.5?
The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.
Which costs less, Kimi K3 or MiMo-V2.5?
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, Kimi K3 or MiMo-V2.5?
Kimi K3 has the larger documented context window: 1.05M, compared with 1M.
Related comparisons
Last updated August 19, 2026
Watch Kimi K3 vs MiMo-V2.5
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.