Agentic
Directional only- Kimi K2.5
- 46.5
- Estimated · #78/153
- MiniMax M2.7
- 41.1
- Estimated · #110/153
- Basis
- BenchAlign lane · 14 vs 7 public rows
- Reading
- Directional only
Kimi K2.5 vs MiniMax M2.7
Updated September 15, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
MiniMax M2.7 has the higher public score estimate, 55.14 versus 54.14, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
12 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 K2.5
Kimi K2.5 has the larger documented context window.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
MiniMax M2.7
MiniMax M2.7 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
MiniMax M2.7
MiniMax M2.7 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
Kimi K2.5 and MiniMax M2.7 are scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Kimi K2.5 and MiniMax M2.7 are scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limited
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. MiniMax M2.7 does not fit this workload in one request. Kimi K2.5 has no published cached-input rate, so cached tokens use its listed input rate. MiniMax M2.7 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
33 Kimi K2.5 only
12 Shared
11 MiniMax M2.7 only
- Shared results
- 12
- Kimi K2.5 only
- 33
- MiniMax M2.7 only
- 11
- Like-for-like categories
- 0 / 8
4 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
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
Directional only- Kimi K2.5
- 50.6
- Estimated · #56/152
- MiniMax M2.7
- 48.6
- Estimated · #68/152
- Basis
- BenchAlign lane · 8 vs 11 public rows
- Reading
- Directional only
Knowledge
Directional only- Kimi K2.5
- 52.1
- Estimated · #69/183
- MiniMax M2.7
- 48.7
- Supported · #90/183
- Basis
- BenchAlign lane · 6 vs 4 public rows
- Reading
- Directional only
Instruction following
Directional only- Kimi K2.5
- 85.8
- #41/123
- MiniMax M2.7
- 93.0
- #10/123
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- Kimi K2.5
- 54.1
- Unranked · 3 rankable rows
- MiniMax M2.7
- 74.8
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Kimi K2.5
- 62.4
- #5/7
- MiniMax M2.7
- Not ranked
- Basis
- Provisional lane · 4 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K2.5
- 38.2
- #8/12
- MiniMax M2.7
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Kimi K2.5
- 65.7
- #24/48
- MiniMax M2.7
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
| Category | Kimi K2.5 | MiniMax M2.7 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 46.5Estimated · #78/153 | 41.1Estimated · #110/153 | Directional onlyBenchAlign lane · 14 vs 7 public rows | Directional only |
| Coding | 50.6Estimated · #56/152 | 48.6Estimated · #68/152 | Directional onlyBenchAlign lane · 8 vs 11 public rows | Directional only |
| Knowledge | 52.1Estimated · #69/183 | 48.7Supported · #90/183 | Directional onlyBenchAlign lane · 6 vs 4 public rows | Directional only |
| Instruction following | 85.8#41/123 | 93.0#10/123 | Directional onlyProvisional lane · 0 vs 0 weighted rows | Directional only |
| Reasoning | 54.1Unranked · 3 rankable rows | 74.8Unranked · 2 rankable rows | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Math | 62.4#5/7 | Not ranked | Not comparableProvisional lane · 4 vs 0 weighted rows | Not comparable |
| Multilingual | 38.2#8/12 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Multimodal | 65.7#24/48 | 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.
SWE-Rebench
Coding
Terminal-Bench 2.0
Agentic
SWE-bench Pro
Coding
SWE Multilingual
Coding
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
- MiniMax M2.7
- $0.0009
- Fits in one request
MiniMax M2.7 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Kimi K2.5
- $0.039
- Fits in one request
- MiniMax M2.7
- $0.0186
- Fits in one request
MiniMax M2.7 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 K2.5
- $0.162
- Fits in one request
- Cached input priced at the published list-input rate
- MiniMax M2.7
- $0.078
- Does not fit in one request
- Cached input priced at the published list-input rate
MiniMax M2.7 does not fit this workload in one request. Kimi K2.5 has no published cached-input rate, so cached tokens use its listed input rate. MiniMax M2.7 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.
SpecificationKimi K2.5MiniMax M2.7
Context window
Maximum documented context; output-token limits may be lower.
Kimi K2.5
256K
MiniMax M2.7
200K
API model ID
Kimi K2.5
Not sourced
MiniMax M2.7
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
MiniMax M2.7
Not published
Documented inputs
Kimi K2.5
Not sourced
MiniMax M2.7
Not sourced
Documented outputs
Kimi K2.5
Not sourced
MiniMax M2.7
Not sourced
Provider availability
Kimi K2.5
Not sourced
MiniMax M2.7
Not sourced
Reasoning profile
Kimi K2.5
Non-Reasoning
MiniMax M2.7
Non-Reasoning
Weight access
Kimi K2.5
Open Weight
MiniMax M2.7
Open Weight
License
Kimi K2.5
Open Weight
MiniMax M2.7
Open Weight
Release date
Kimi K2.5
2026-02-01
MiniMax M2.7
2026-03-18
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
- MiniMax M2.7 has the higher public score estimate, 55.14 versus 54.14, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.039 vs $0.0186. Cache-heavy agent loop: $0.162 vs $0.078.
- Context tradeoff
- Kimi K2.5 has the larger documented window (256K).
Run the same representative tasks against both endpoints before changing production traffic.
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$5,221
Break-even132M/day
MiniMax M2.7
API / mo$1,125
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 evidence56 rows
Agentic
Terminal-Bench 2.0
Kimi K2.550.8%SourceMiniMax M2.757%SourceMiniMax M2.7 leads this result
BrowseComp
Kimi K2.560.6%SourceMiniMax M2.7—Not directly comparable
Claw-Eval
Shared sourceKimi K2.552.3%MiniMax M2.748.7%Kimi K2.5 leads this result
QwenClawBench
Kimi K2.554.3%SourceMiniMax M2.7—Not directly comparable
τ³-bench results
Kimi K2.565.7%SourceMiniMax M2.7—Not directly comparable
DeepSearchQA
Kimi K2.577.1%SourceMiniMax M2.7—Not directly comparable
DeepPlanning
Kimi K2.514.4%SourceMiniMax M2.7—Not directly comparable
Toolathlon
Kimi K2.527.8%SourceMiniMax M2.746.3%SourceMiniMax M2.7 leads this result
MCP Atlas
Kimi K2.529.5%SourceMiniMax M2.7—Not directly comparable
MCP-Tasks
Kimi K2.559.1%SourceMiniMax M2.7—Not directly comparable
WideResearch
Kimi K2.572.7%SourceMiniMax M2.7—Not directly comparable
Gert Labs
Shared sourceKimi K2.545.88%MiniMax M2.740.40%Kimi K2.5 leads this result
ResearchClawBench
Kimi K2.514.0%SourceMiniMax M2.7—Not directly comparable
JobBench
Kimi K2.58.7%SourceMiniMax M2.7—Not directly comparable
MLE-Bench Lite
Kimi K2.5—MiniMax M2.766.6%SourceNot directly comparable
MM-ClawBench
Kimi K2.5—MiniMax M2.762.7%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Kimi K2.5—MiniMax M2.748.7%SourceNot directly comparable
Coding
SWE-bench Verified
Kimi K2.576.8%SourceMiniMax M2.7—Not directly comparable
SWE-bench Verified*
Shared sourceKimi K2.570.8%MiniMax M2.775.4%MiniMax M2.7 leads this result
LiveCodeBench v6
Kimi K2.585.0%SourceMiniMax M2.7—Not directly comparable
SWE-bench Pro
Kimi K2.550.7%SourceMiniMax M2.756.2%SourceMiniMax M2.7 leads this result
SWE Multilingual
Kimi K2.573%SourceMiniMax M2.776.5%SourceMiniMax M2.7 leads this result
SWE-Rebench
Kimi K2.558.5%SourceMiniMax M2.751.9%SourceKimi K2.5 leads this result
React Native Evals
Shared sourceKimi K2.577.2%MiniMax M2.771.4%Kimi K2.5 leads this result
SciCode
Kimi K2.548.7%SourceMiniMax M2.7—Not directly comparable
Multi-SWE Bench
Kimi K2.5—MiniMax M2.752.7%SourceNot directly comparable
VIBE-Pro
Kimi K2.5—MiniMax M2.755.6%SourceNot directly comparable
NL2Repo
Kimi K2.5—MiniMax M2.739.8%SourceNot directly comparable
Vibe Code Bench
Kimi K2.5—MiniMax M2.727.04%SourceNot directly comparable
LiveCodeBench (Vals)
Kimi K2.5—MiniMax M2.779.9%SourceNot directly comparable
SWE-bench (Vals)
Kimi K2.5—MiniMax M2.773.8%SourceNot directly comparable
Reasoning
LongBench v2
Kimi K2.561%SourceMiniMax M2.7—Not directly comparable
Knowledge
GPQA
Kimi K2.587.6%SourceMiniMax M2.7—Not directly comparable
GPQA-D
Kimi K2.587.6%SourceMiniMax M2.787.0%SourceKimi K2.5 leads this result
SuperGPQA
Kimi K2.569.2%SourceMiniMax M2.7—Not directly comparable
MMLU-Pro
Kimi K2.587.1%SourceMiniMax M2.7—Not directly comparable
MMLU-Pro (Arcee)
Shared sourceKimi K2.587.1%MiniMax M2.780.8%Kimi K2.5 leads this result
HLE
Kimi K2.530.1%SourceMiniMax M2.7—Not directly comparable
GPQA Diamond (Vals)
Kimi K2.5—MiniMax M2.786.6%SourceNot directly comparable
MMLU-Pro (Vals)
Kimi K2.5—MiniMax M2.780.4%SourceNot directly comparable
Math
AIME 2025
Kimi K2.596.1%SourceMiniMax M2.7—Not directly comparable
AIME26
Kimi K2.595.8%SourceMiniMax M2.7—Not directly comparable
AIME25 (Arcee)
Shared sourceKimi K2.596.3%MiniMax M2.780.0%Kimi K2.5 leads this result
HMMT Feb 2025
Kimi K2.595.4%SourceMiniMax M2.7—Not directly comparable
HMMT Nov 2025
Kimi K2.591.1%SourceMiniMax M2.7—Not directly comparable
HMMT Feb 2026
Kimi K2.587.1%SourceMiniMax M2.7—Not directly comparable
MMAnswerBench
Kimi K2.581.8%SourceMiniMax M2.7—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Kimi K2.527.900%SourceMiniMax M2.7—Not directly comparable
FrontierMath v2 (Tier 4)
Kimi K2.54.200%SourceMiniMax M2.7—Not directly comparable
Multilingual
Multimodal
Instruction following
IFEval
Kimi K2.593.9%SourceMiniMax M2.7—Not directly comparable
Frequently asked questions
Which is better, Kimi K2.5 or MiniMax M2.7?
MiniMax M2.7 has the higher public score estimate, 55.14 versus 54.14, 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 MiniMax M2.7?
Kimi K2.5 scores higher for coding on the public lane, 50.6 to 48.6. Kimi K2.5 and MiniMax M2.7 are 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 MiniMax M2.7?
Kimi K2.5 scores higher for agentic tasks on the public lane, 46.5 to 41.1. Kimi K2.5 and MiniMax M2.7 are 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 MiniMax M2.7?
For the stated presets, chat costs $0.0021 on Kimi K2.5 and $0.0009 on MiniMax M2.7; repository review costs $0.039 and $0.0186; the cache-heavy agent loop costs $0.162 and $0.078. MiniMax M2.7 does not fit this workload in one request. Kimi K2.5 has no published cached-input rate, so cached tokens use its listed input rate. MiniMax M2.7 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Kimi K2.5 or MiniMax M2.7?
Kimi K2.5 has the larger documented context window: 256K, compared with 200K.
Related comparisons
Last updated September 15, 2026
Watch Kimi K2.5 vs MiniMax M2.7
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.