Long documents
Prompts that approach the documented context limit
MiMo-V2.5-Pro
MiMo-V2.5-Pro has the larger documented context window.
Confidence: documented
Kimi K2.5 vs MiMo-V2.5-Pro
Updated October 20, 2026. Rank cannot separate these two. Price, access, and your workload decide. Public scores include evidence status and uncertainty.
Share or export
Decision reading
Kimi K2.5 has the higher public score estimate, 53.03 versus 53.02, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 7 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 7
- Kimi K2.5 only
- 61
- MiMo-V2.5-Pro only
- 7
- Like-for-like categories
- 0 / 8
Supported: Kimi K2.5 and MiMo-V2.5-ProHow the comparison works
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.
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Kimi K2.5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limitedAgentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Kimi K2.5 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limitedChat 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-ratesCache-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: rate-fallbackRepository 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
Which one for a specific job
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
38.5Kimi K2.543.2MiMo-V2.5-Pro
Directional only · BenchAlign v5.8
MiMo-V2.5-Pro has the higher coding point estimate. Conditional score ranges do not establish rank confidence.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
61 Kimi K2.5 only
7 Shared
7 MiMo-V2.5-Pro only
4 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
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.
HLEKnowledge
Normalized gap 17.9Kimi K2.5:30.1%MiMo-V2.5-Pro:48%Terminal-Bench 2.0Agentic
Normalized gap 17.6Kimi K2.5:50.8%MiMo-V2.5-Pro:68.4%SWE-bench ProCoding
Normalized gap 6.5Kimi K2.5:50.7%MiMo-V2.5-Pro:57.2%
Bars run 0–100 on each benchmark’s normalized display scale
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign v5.8 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.
Agentic
Directional only- Kimi K2.5
- 40.9
- Estimated · #67/131
- MiMo-V2.5-Pro
- 37.8
- Supported · #74/131
- Basis
- BenchAlign v5.8 lane · 16 vs 6 public rows
- Reading
- Directional only
Coding
Directional only- Kimi K2.5
- 38.5
- Estimated · #69/152
- MiMo-V2.5-Pro
- 43.2
- Supported · #58/152
- Basis
- BenchAlign v5.8 lane · 10 vs 4 public rows
- Reading
- Directional only
Knowledge
Directional only- Kimi K2.5
- 51.4
- Estimated · #78/187
- MiMo-V2.5-Pro
- 52.4
- Supported · #73/187
- Basis
- BenchAlign v5.8 lane · 8 vs 4 public rows
- Reading
- Directional only
Instruction following
Directional only- Kimi K2.5
- 85.8
- #48/149
- MiMo-V2.5-Pro
- 93.8
- #7/149
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- Kimi K2.5
- 71.3
- Unranked · 3 rankable rows
- MiMo-V2.5-Pro
- 79.8
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Kimi K2.5
- 70.4
- #33/55
- MiMo-V2.5-Pro
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K2.5
- 87.3
- #8/17
- MiMo-V2.5-Pro
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Kimi K2.5
- 83.3
- Unranked · 7 rankable rows
- MiMo-V2.5-Pro
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
| Category | Kimi K2.5 | MiMo-V2.5-Pro | Basis | Reading |
|---|---|---|---|---|
| Agentic | 40.9Estimated · #67/131 | 37.8Supported · #74/131 | Directional onlyBenchAlign v5.8 lane · 16 vs 6 public rows | Directional only |
| Coding | 38.5Estimated · #69/152 | 43.2Supported · #58/152 | Directional onlyBenchAlign v5.8 lane · 10 vs 4 public rows | Directional only |
| Knowledge | 51.4Estimated · #78/187 | 52.4Supported · #73/187 | Directional onlyBenchAlign v5.8 lane · 8 vs 4 public rows | Directional only |
| Instruction following | 85.8#48/149 | 93.8#7/149 | Directional onlyProvisional lane · 0 vs 0 weighted rows | Directional only |
| Reasoning | 71.3Unranked · 3 rankable rows | 79.8Unranked · 2 rankable rows | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Multimodal | 70.4#33/55 | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Multilingual | 87.3#8/17 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Math | 83.3Unranked · 7 rankable rows | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Supported evidence per lane · bars run 0–100Methodology
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 K2.5
- $0.0021
- Fits in one request
- MiMo-V2.5-Pro
- API rate not published
- Fits in one request
MiMo-V2.5-Pro has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Kimi K2.5
- $0.039
- Fits in one request
- MiMo-V2.5-Pro
- API rate not published
- Fits in one request
MiMo-V2.5-Pro has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Kimi K2.5
- $0.162
- Fits in one request
- Cached input priced at the published list-input rate
- MiMo-V2.5-Pro
- API rate not published
- Fits in one request
- Cached-input rate unavailable
Kimi K2.5 has no published cached-input rate, so cached tokens use its listed input rate. MiMo-V2.5-Pro has no comparable published API token rate.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationKimi K2.5MiMo-V2.5-Pro
Context window
Maximum documented context; output-token limits may be lower.
Kimi K2.5
256K
MiMo-V2.5-Pro
1M
API model ID
Kimi K2.5
Not sourced
MiMo-V2.5-Pro
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Kimi K2.5
Not published
MiMo-V2.5-Pro
No comparable hosted API rate
Documented inputs
Kimi K2.5
Not sourced
MiMo-V2.5-Pro
Not sourced
Documented outputs
Kimi K2.5
Not sourced
MiMo-V2.5-Pro
Not sourced
Provider availability
Kimi K2.5
Not sourced
MiMo-V2.5-Pro
Not sourced
Reasoning profile
Kimi K2.5
Non-Reasoning
MiMo-V2.5-Pro
Reasoning
Weight access
Kimi K2.5
Open Weight
MiMo-V2.5-Pro
Proprietary
License
Kimi K2.5
Open Weight
MiMo-V2.5-Pro
Proprietary
Release date
Kimi K2.5
2026-01-27
MiMo-V2.5-Pro
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 K2.5 has the higher public score estimate, 53.03 versus 53.02, but the 90% score intervals overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- MiMo-V2.5-Pro has the larger documented window (1M).
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Kimi K2.5 or MiMo-V2.5-Pro?
Kimi K2.5 has the higher public score estimate, 53.03 versus 53.02, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, Kimi K2.5 or MiMo-V2.5-Pro?
MiMo-V2.5-Pro scores higher for coding on the public lane, 43.2 to 38.5. Kimi K2.5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which is better for agentic tasks, Kimi K2.5 or MiMo-V2.5-Pro?
Kimi K2.5 scores higher for agentic tasks on the public lane, 40.9 to 37.8. Kimi K2.5 is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which costs less, Kimi K2.5 or MiMo-V2.5-Pro?
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 K2.5 or MiMo-V2.5-Pro?
MiMo-V2.5-Pro has the larger documented context window: 1M, compared with 256K.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Kimi K2.5
API / mo$2,700
Self-host / mo$7,358
Break-even186M/day
MiMo-V2.5-Pro
API / mo$0
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence75 rows
Agentic
Terminal-Bench 2.0
Kimi K2.550.8%SourceMiMo-V2.5-Pro68.4%SourceMiMo-V2.5-Pro leads this result
BrowseComp
Kimi K2.560.6%SourceMiMo-V2.5-Pro—Not directly comparable
OSWorld-Verified
Kimi K2.563.3%SourceMiMo-V2.5-Pro—Not directly comparable
Claw-Eval
Shared sourceKimi K2.552.3%MiMo-V2.5-Pro63.8%MiMo-V2.5-Pro leads this result
QwenClawBench
Kimi K2.554.3%SourceMiMo-V2.5-Pro—Not directly comparable
τ³-bench results
Kimi K2.565.7%SourceMiMo-V2.5-Pro72.9%SourceMiMo-V2.5-Pro leads this result
DeepSearchQA
Kimi K2.577.1%SourceMiMo-V2.5-Pro—Not directly comparable
DeepPlanning
Kimi K2.514.4%SourceMiMo-V2.5-Pro—Not directly comparable
Toolathlon
Kimi K2.527.8%SourceMiMo-V2.5-Pro—Not directly comparable
MCP Atlas
Kimi K2.529.5%SourceMiMo-V2.5-Pro—Not directly comparable
MCP-Tasks
Kimi K2.559.1%SourceMiMo-V2.5-Pro—Not directly comparable
WideResearch
Kimi K2.572.7%SourceMiMo-V2.5-Pro—Not directly comparable
BFCL v4
Kimi K2.568.3%SourceMiMo-V2.5-Pro—Not directly comparable
Gert Labs
Shared sourceKimi K2.545.88%MiMo-V2.5-Pro62.70%MiMo-V2.5-Pro leads this result
ResearchClawBench
Kimi K2.514.0%SourceMiMo-V2.5-Pro—Not directly comparable
JobBench
Kimi K2.58.7%SourceMiMo-V2.5-Pro—Not directly comparable
Terminal-Bench 2.1 (Vals)
Kimi K2.5—MiMo-V2.5-Pro57.3%SourceNot directly comparable
MazeBench
Kimi K2.5—MiMo-V2.5-Pro0%SourceNot directly comparable
Coding
SWE-bench Verified
Kimi K2.576.8%SourceMiMo-V2.5-Pro—Not directly comparable
SWE-bench Verified*
Kimi K2.570.8%SourceMiMo-V2.5-Pro—Not directly comparable
LiveCodeBench v6
Kimi K2.585.0%SourceMiMo-V2.5-Pro—Not directly comparable
SWE-bench Pro
Kimi K2.550.7%SourceMiMo-V2.5-Pro57.2%SourceMiMo-V2.5-Pro leads this result
SWE Multilingual
Kimi K2.573%SourceMiMo-V2.5-Pro—Not directly comparable
SWE-Rebench
Kimi K2.558.5%SourceMiMo-V2.5-Pro—Not directly comparable
React Native Evals
Kimi K2.577.2%SourceMiMo-V2.5-Pro—Not directly comparable
NL2Repo
Kimi K2.532%SourceMiMo-V2.5-Pro—Not directly comparable
SciCode
Kimi K2.548.7%SourceMiMo-V2.5-Pro—Not directly comparable
Terminal-Bench 2.0
Kimi K2.550.8%SourceMiMo-V2.5-Pro68.4%SourceMiMo-V2.5-Pro leads this result
LiveCodeBench (Vals)
Kimi K2.5—MiMo-V2.5-Pro81.4%SourceNot directly comparable
SWE-bench (Vals)
Kimi K2.5—MiMo-V2.5-Pro74.0%SourceNot directly comparable
Reasoning
LongBench v2
Kimi K2.561%SourceMiMo-V2.5-Pro—Not directly comparable
Multimodal
MMMU
Kimi K2.584.3%SourceMiMo-V2.5-Pro—Not directly comparable
MMMU-Pro
Kimi K2.578.5%SourceMiMo-V2.5-Pro—Not directly comparable
RealWorldQA
Kimi K2.581.0%SourceMiMo-V2.5-Pro—Not directly comparable
OmniDocBench 1.5
Kimi K2.588.8%SourceMiMo-V2.5-Pro—Not directly comparable
Video-MME
Kimi K2.587.4%SourceMiMo-V2.5-Pro—Not directly comparable
Video-MME (with subtitle)
Kimi K2.587.4%SourceMiMo-V2.5-Pro—Not directly comparable
Video-MME (w/o subtitle)
Kimi K2.583.2%SourceMiMo-V2.5-Pro—Not directly comparable
MathVision
Kimi K2.584.2%SourceMiMo-V2.5-Pro—Not directly comparable
We-Math
Kimi K2.584.7%SourceMiMo-V2.5-Pro—Not directly comparable
DynaMath
Kimi K2.584.4%SourceMiMo-V2.5-Pro—Not directly comparable
SimpleVQA
Kimi K2.571.2%SourceMiMo-V2.5-Pro—Not directly comparable
MMLongBench-Doc
Kimi K2.558.5%SourceMiMo-V2.5-Pro—Not directly comparable
CC-OCR
Kimi K2.579.7%SourceMiMo-V2.5-Pro—Not directly comparable
AI2D_TEST
Kimi K2.590.8%SourceMiMo-V2.5-Pro—Not directly comparable
CountBench
Kimi K2.594.1%SourceMiMo-V2.5-Pro—Not directly comparable
RefCOCO (avg)
Kimi K2.587.8%SourceMiMo-V2.5-Pro—Not directly comparable
MMVU
Kimi K2.580.4%SourceMiMo-V2.5-Pro—Not directly comparable
VideoMMMU
Kimi K2.586.6%SourceMiMo-V2.5-Pro—Not directly comparable
MLVU (M-Avg)
Kimi K2.585.0%SourceMiMo-V2.5-Pro—Not directly comparable
CharXiv
Kimi K2.577.5%SourceMiMo-V2.5-Pro—Not directly comparable
Knowledge
GPQA
Kimi K2.587.6%SourceMiMo-V2.5-Pro—Not directly comparable
GPQA-D
Kimi K2.587.6%SourceMiMo-V2.5-Pro—Not directly comparable
SuperGPQA
Kimi K2.569.2%SourceMiMo-V2.5-Pro—Not directly comparable
MMLU-Pro
Kimi K2.587.1%SourceMiMo-V2.5-Pro—Not directly comparable
MMLU-Pro (Arcee)
Kimi K2.587.1%SourceMiMo-V2.5-Pro—Not directly comparable
MMLU-Redux
Kimi K2.594.5%SourceMiMo-V2.5-Pro—Not directly comparable
C-Eval
Kimi K2.594%SourceMiMo-V2.5-Pro—Not directly comparable
HLE
Kimi K2.530.1%SourceMiMo-V2.5-Pro48%SourceMiMo-V2.5-Pro leads this result
HLE w/o tools
Kimi K2.5—MiMo-V2.5-Pro34%SourceNot directly comparable
GPQA Diamond (Vals)
Kimi K2.5—MiMo-V2.5-Pro82.6%SourceNot directly comparable
MMLU-Pro (Vals)
Kimi K2.5—MiMo-V2.5-Pro84.6%SourceNot directly comparable
Multilingual
MMLU-ProX
Kimi K2.582.3%SourceMiMo-V2.5-Pro—Not directly comparable
NOVA-63
Kimi K2.556.0%SourceMiMo-V2.5-Pro—Not directly comparable
INCLUDE
Kimi K2.583.3%SourceMiMo-V2.5-Pro—Not directly comparable
PolyMath
Kimi K2.543.1%SourceMiMo-V2.5-Pro—Not directly comparable
MAXIFE
Kimi K2.572.8%SourceMiMo-V2.5-Pro—Not directly comparable
Instruction following
IFEval
Kimi K2.593.9%SourceMiMo-V2.5-Pro—Not directly comparable
Math
AIME 2025
Kimi K2.596.1%SourceMiMo-V2.5-Pro—Not directly comparable
AIME26
Kimi K2.595.8%SourceMiMo-V2.5-Pro—Not directly comparable
AIME25 (Arcee)
Kimi K2.596.3%SourceMiMo-V2.5-Pro—Not directly comparable
HMMT Feb 2025
Kimi K2.595.4%SourceMiMo-V2.5-Pro—Not directly comparable
HMMT Nov 2025
Kimi K2.591.1%SourceMiMo-V2.5-Pro—Not directly comparable
HMMT Feb 2026
Kimi K2.587.1%SourceMiMo-V2.5-Pro—Not directly comparable
MMAnswerBench
Kimi K2.581.8%SourceMiMo-V2.5-Pro—Not directly comparable
75 public results · 7 shared
Watch Kimi K2.5 vs MiMo-V2.5-Pro
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 5,600+ readers.
Last updated October 20, 2026