Agentic
Like-for-like- Kimi K2.6
- 73.5
- MiniMax M3
- 72.3
- Weighted basis
- 3 vs 3 rows
- Reading
- Kimi K2.6 leads
Kimi K2.6 vs MiniMax M3
Updated August 30, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
MiniMax M3 has the higher public score estimate, 68.73 versus 60.11, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
11 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.
Agentic work
Tool use, computer use, and multi-step task completion
Kimi K2.6
Kimi K2.6 leads on the same 3 weighted benchmark rows.
Confidence: stronger
Long documents
Prompts that approach the documented context limit
MiniMax M3
MiniMax M3 has the larger documented context window.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
MiniMax M3
MiniMax M3 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
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
MiniMax M3
MiniMax M3 has the lower estimated token cost for this stated workload. Kimi K2.6 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
MiniMax M3
MiniMax M3 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
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
21 Kimi K2.6 only
11 Shared
11 MiniMax M3 only
- Shared results
- 11
- Kimi K2.6 only
- 21
- MiniMax M3 only
- 11
- Like-for-like categories
- 1 / 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.
Coding
Directional only- Kimi K2.6
- 64.4
- MiniMax M3
- 72.2
- Weighted basis
- 3 vs 2 rows
- Reading
- Directional only
Multimodal
Directional only- Kimi K2.6
- 79.8
- MiniMax M3
- 64.9
- Weighted basis
- 2 vs 2 rows
- Reading
- Directional only
Reasoning
Not comparable- Kimi K2.6
- Not measured
- MiniMax M3
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Knowledge
Not comparable- Kimi K2.6
- 42.2
- MiniMax M3
- Not measured
- Weighted basis
- 2 vs 0 rows
- Reading
- Not comparable
Math
Not comparable- Kimi K2.6
- 67.1
- MiniMax M3
- 85.7
- Weighted basis
- 4 vs 1 rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K2.6
- Not measured
- MiniMax M3
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Instruction following
Not comparable- Kimi K2.6
- Not measured
- MiniMax M3
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | Kimi K2.6 | MiniMax M3 | Weighted basis | Reading |
|---|---|---|---|---|
| Agentic | 73.5 | 72.3 | Like-for-like3 vs 3 rows | Kimi K2.6 leads |
| Coding | 64.4 | 72.2 | Directional only3 vs 2 rows | Directional only |
| Multimodal | 79.8 | 64.9 | Directional only2 vs 2 rows | Directional only |
| Reasoning | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Knowledge | 42.2 | Not measured | Not comparable2 vs 0 rows | Not comparable |
| Math | 67.1 | 85.7 | Not comparable4 vs 1 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.
OSWorld-Verified
Agentic
MMMU-Pro
Multimodal
Terminal-Bench 2.0
Agentic
SWE-bench Pro
Coding
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
- Kimi K2.6
- $0.00295
- Fits in one request
- MiniMax M3
- $0.0009
- Fits in one request
MiniMax M3 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Kimi K2.6
- $0.0595
- Fits in one request
- MiniMax M3
- $0.0186
- Fits in one request
MiniMax M3 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 K2.6
- $0.249
- Fits in one request
- Cached input priced at the published list-input rate
- MiniMax M3
- $0.03
- Fits in one request
MiniMax M3 has the lower modeled cost
Kimi K2.6 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 K2.6MiniMax M3
Context window
Maximum documented context; output-token limits may be lower.
Kimi K2.6
256K
MiniMax M3
1M
API model ID
Kimi K2.6
Not sourced
MiniMax M3
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Kimi K2.6
Not published
MiniMax M3
$0.06 per 1M cached input tokens
Documented inputs
Kimi K2.6
Not sourced
MiniMax M3
Not sourced
Documented outputs
Kimi K2.6
Not sourced
MiniMax M3
Not sourced
Provider availability
Kimi K2.6
Not sourced
MiniMax M3
Not sourced
Reasoning profile
Kimi K2.6
Reasoning
MiniMax M3
Non-Reasoning
Weight access
Kimi K2.6
Open Weight
MiniMax M3
Open Weight
License
Kimi K2.6
Open Weight
MiniMax M3
Open Weight
Release date
Kimi K2.6
2026-04-20
MiniMax M3
2026-06-01
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 M3 has the higher public score estimate, 68.73 versus 60.11, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.0595 vs $0.0186. Cache-heavy agent loop: $0.249 vs $0.03.
- Context tradeoff
- MiniMax M3 has the larger documented window (1M).
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 K2.6
API / mo$3,713
Self-host / mo$18,221
Break-even326M/day
MiniMax M3
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 evidence43 rows
Agentic
Terminal-Bench 2.0
Kimi K2.666.7%SourceMiniMax M366%SourceKimi K2.6 leads this result
BrowseComp
Kimi K2.683.2%SourceMiniMax M383.5%SourceMiniMax M3 leads this result
OSWorld-Verified
Kimi K2.673.1%SourceMiniMax M370.1%SourceKimi K2.6 leads this result
Toolathlon
Kimi K2.650%SourceMiniMax M3—Not directly comparable
MCP Atlas
Kimi K2.655.9%SourceMiniMax M374.2%SourceMiniMax M3 leads this result
Claw-Eval
Kimi K2.662.3%SourceMiniMax M374.5%SourceMiniMax M3 leads this result
DeepSearchQA
Kimi K2.692.5%SourceMiniMax M3—Not directly comparable
WideResearch
Kimi K2.680.8%SourceMiniMax M3—Not directly comparable
Gert Labs
Kimi K2.656.82%SourceMiniMax M3—Not directly comparable
ResearchClawBench
Shared sourceKimi K2.618.0%MiniMax M319.8%MiniMax M3 leads this result
OSWorld 2.0
Shared sourceKimi K2.64.6%MiniMax M34.6%Tie
BankerToolBench
Kimi K2.6—MiniMax M376.1%SourceNot directly comparable
Coding
SWE-bench Verified
Kimi K2.680.2%SourceMiniMax M380.5%SourceMiniMax M3 leads this result
LiveCodeBench v6
Kimi K2.689.6%SourceMiniMax M3—Not directly comparable
SWE-bench Pro
Kimi K2.658.6%SourceMiniMax M359%SourceMiniMax M3 leads this result
SWE Multilingual
Kimi K2.676.7%SourceMiniMax M3—Not directly comparable
SciCode
Kimi K2.652.2%SourceMiniMax M3—Not directly comparable
Terminal-Bench 2.0
Kimi K2.666.7%SourceMiniMax M366.0%SourceKimi K2.6 leads this result
Vibe Code Bench
Kimi K2.637.89%SourceMiniMax M3—Not directly comparable
cursorBench31
Kimi K2.647.6%SourceMiniMax M3—Not directly comparable
NL2Repo
Kimi K2.6—MiniMax M342.1%SourceNot directly comparable
VIBE V2
Kimi K2.6—MiniMax M350.1%SourceNot directly comparable
SVG-Bench
Kimi K2.6—MiniMax M363.7%SourceNot directly comparable
KernelBench Hard
Kimi K2.6—MiniMax M328.8%SourceNot directly comparable
OpenHarmony Bench
Kimi K2.6—MiniMax M348.4%SourceNot directly comparable
Knowledge
Math
AIME26
Kimi K2.696.4%SourceMiniMax M3—Not directly comparable
HMMT Feb 2026
Kimi K2.692.7%SourceMiniMax M3—Not directly comparable
MMAnswerBench
Kimi K2.686.0%SourceMiniMax M3—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Kimi K2.638.966%SourceMiniMax M3—Not directly comparable
FrontierMath v2 (Tier 4)
Kimi K2.614.580%SourceMiniMax M3—Not directly comparable
USAMO 2026
Kimi K2.6—MiniMax M385.7%SourceNot directly comparable
Multimodal
MMMU-Pro
Kimi K2.679.4%SourceMiniMax M378.1%SourceKimi K2.6 leads this result
MMMU-Pro w/ Python
Kimi K2.680.1%SourceMiniMax M3—Not directly comparable
CharXiv
Kimi K2.680.4%SourceMiniMax M3—Not directly comparable
MathVision
Kimi K2.687.4%SourceMiniMax M3—Not directly comparable
V*
Kimi K2.696.9%SourceMiniMax M3—Not directly comparable
OfficeQA Pro
Kimi K2.6—MiniMax M345.1%SourceNot directly comparable
OmniDocBench 1.5
Kimi K2.6—MiniMax M391.6%SourceNot directly comparable
VideoMMMU
Kimi K2.6—MiniMax M384.6%SourceNot directly comparable
Video-MME (with subtitle)
Kimi K2.6—MiniMax M385.4%SourceNot directly comparable
Frequently asked questions
Which is better, Kimi K2.6 or MiniMax M3?
MiniMax M3 has the higher public score estimate, 68.73 versus 60.11, 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 K2.6 or MiniMax M3?
The current coding 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 is better for agentic tasks, Kimi K2.6 or MiniMax M3?
Kimi K2.6 leads the like-for-like agentic tasks comparison across 3 shared weighted benchmark rows.
Which costs less, Kimi K2.6 or MiniMax M3?
For the stated presets, chat costs $0.00295 on Kimi K2.6 and $0.0009 on MiniMax M3; repository review costs $0.0595 and $0.0186; the cache-heavy agent loop costs $0.249 and $0.03. Kimi K2.6 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Kimi K2.6 or MiniMax M3?
MiniMax M3 has the larger documented context window: 1M, compared with 256K.
Related comparisons
Last updated August 30, 2026
Watch Kimi K2.6 vs MiniMax M3
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.