Agentic
Like-for-like- Grok 4.3
- 26.9
- Supported · #144/152
- Kimi K2.6
- 46.7
- Supported · #74/152
- Basis
- BenchAlign lane · 3 vs 12 public rows
- Reading
- Kimi K2.6 leads · intervals overlap
Grok 4.3 vs Kimi K2.6
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Kimi K2.6 has the higher public score estimate, 65.39 versus 59.73, 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.
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.
Coding work
Code generation, repair, and software-engineering tasks
Kimi K2.6
Kimi K2.6 leads on the public coding lane, 50.8 to 33.6, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Kimi K2.6
Kimi K2.6 leads on the public agentic lane, 46.7 to 26.9, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Long documents
Prompts that approach the documented context limit
Grok 4.3
Grok 4.3 has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Chat turn cost
1K fresh input + 500 output tokens
Grok 4.3
Grok 4.3 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Grok 4.3
Grok 4.3 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
Kimi K2.6
Kimi K2.6 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
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.
7 Shared
30 Kimi K2.6 only
- Shared results
- 7
- Grok 4.3 only
- 0
- Kimi K2.6 only
- 30
- Like-for-like categories
- 3 / 8
1 category rests 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
Like-for-like- Grok 4.3
- 33.6
- Supported · #131/151
- Kimi K2.6
- 50.8
- Supported · #52/151
- Basis
- BenchAlign lane · 2 vs 10 public rows
- Reading
- Kimi K2.6 leads · intervals overlap
Knowledge
Like-for-like- Grok 4.3
- 57.0
- Supported · #44/183
- Kimi K2.6
- 61.5
- Supported · #32/183
- Basis
- BenchAlign lane · 2 vs 5 public rows
- Reading
- Kimi K2.6 leads · intervals overlap
Multimodal
Directional only- Grok 4.3
- 65.6
- #25/48
- Kimi K2.6
- 64.0
- #26/48
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- Grok 4.3
- 65.2
- Unranked · 2 rankable rows
- Kimi K2.6
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Grok 4.3
- Not ranked
- Kimi K2.6
- 71.3
- #1/7
- Basis
- Provisional lane · 0 vs 4 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Grok 4.3
- Not ranked
- Kimi K2.6
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Grok 4.3
- 94.4
- #2/123
- Kimi K2.6
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Grok 4.3 | Kimi K2.6 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 26.9Supported · #144/152 | 46.7Supported · #74/152 | Like-for-likeBenchAlign lane · 3 vs 12 public rows | Kimi K2.6 leads · intervals overlap |
| Coding | 33.6Supported · #131/151 | 50.8Supported · #52/151 | Like-for-likeBenchAlign lane · 2 vs 10 public rows | Kimi K2.6 leads · intervals overlap |
| Knowledge | 57.0Supported · #44/183 | 61.5Supported · #32/183 | Like-for-likeBenchAlign lane · 2 vs 5 public rows | Kimi K2.6 leads · intervals overlap |
| Multimodal | 65.6#25/48 | 64.0#26/48 | Directional onlyProvisional lane · 0 vs 2 weighted rows | Directional only |
| Reasoning | 65.2Unranked · 2 rankable rows | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | 71.3#1/7 | Not comparableProvisional lane · 0 vs 4 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | 94.4#2/123 | Not ranked | Not comparableProvisional lane · 0 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.
LiveCodeBench (Vals)
Coding
MMLU-Pro (Vals)
Knowledge
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
- Grok 4.3
- $0.0025
- Fits in one request
- Kimi K2.6
- $0.00295
- Fits in one request
Grok 4.3 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Grok 4.3
- $0.07
- Fits in one request
- Kimi K2.6
- $0.0595
- Fits in one request
Kimi K2.6 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Grok 4.3
- $0.09
- Fits in one request
- Kimi K2.6
- $0.249
- Fits in one request
- Cached input priced at the published list-input rate
Grok 4.3 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.
SpecificationGrok 4.3Kimi K2.6
Context window
Maximum documented context; output-token limits may be lower.
Grok 4.3
1M
Kimi K2.6
256K
API model ID
Grok 4.3
Not sourced
Kimi K2.6
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Grok 4.3
$0.2 per 1M cached input tokens
Kimi K2.6
Not published
Documented inputs
Grok 4.3
Not sourced
Kimi K2.6
Not sourced
Documented outputs
Grok 4.3
Not sourced
Kimi K2.6
Not sourced
Provider availability
Grok 4.3
Not sourced
Kimi K2.6
Not sourced
Reasoning profile
Grok 4.3
Reasoning
Kimi K2.6
Reasoning
Weight access
Grok 4.3
Proprietary
Kimi K2.6
Open Weight
License
Grok 4.3
Proprietary
Kimi K2.6
Open Weight
Release date
Grok 4.3
2026-04-30
Kimi K2.6
2026-04-20
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 K2.6 has the higher public score estimate, 65.39 versus 59.73, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.07 vs $0.0595. Cache-heavy agent loop: $0.09 vs $0.249.
- Context tradeoff
- Grok 4.3 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.
Grok 4.3
API / mo$2,813
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Kimi K2.6
API / mo$3,713
Self-host / mo$18,221
Break-even326M/day
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence37 rows
Agentic
Gert Labs
Shared sourceGrok 4.343.86%Kimi K2.656.82%Kimi K2.6 leads this result
ResearchClawBench
Shared sourceGrok 4.312.4%Kimi K2.618.0%Kimi K2.6 leads this result
Terminal-Bench 2.1 (Vals)
Grok 4.341.9%SourceKimi K2.653.6%SourceKimi K2.6 leads this result
Terminal-Bench 2.0
Grok 4.3—Kimi K2.666.7%SourceNot directly comparable
BrowseComp
Grok 4.3—Kimi K2.683.2%SourceNot directly comparable
OSWorld-Verified
Grok 4.3—Kimi K2.673.1%SourceNot directly comparable
Toolathlon
Grok 4.3—Kimi K2.650%SourceNot directly comparable
MCP Atlas
Grok 4.3—Kimi K2.655.9%SourceNot directly comparable
Claw-Eval
Grok 4.3—Kimi K2.662.3%SourceNot directly comparable
DeepSearchQA
Grok 4.3—Kimi K2.692.5%SourceNot directly comparable
WideResearch
Grok 4.3—Kimi K2.680.8%SourceNot directly comparable
OSWorld 2.0
Grok 4.3—Kimi K2.64.6%SourceNot directly comparable
Coding
LiveCodeBench (Vals)
Grok 4.384.5%SourceKimi K2.686.8%SourceKimi K2.6 leads this result
SWE-bench (Vals)
Grok 4.371.4%SourceKimi K2.676.2%SourceKimi K2.6 leads this result
SWE-bench Verified
Grok 4.3—Kimi K2.680.2%SourceNot directly comparable
LiveCodeBench v6
Grok 4.3—Kimi K2.689.6%SourceNot directly comparable
SWE-bench Pro
Grok 4.3—Kimi K2.658.6%SourceNot directly comparable
SWE Multilingual
Grok 4.3—Kimi K2.676.7%SourceNot directly comparable
SciCode
Grok 4.3—Kimi K2.652.2%SourceNot directly comparable
Terminal-Bench 2.0
Grok 4.3—Kimi K2.666.7%SourceNot directly comparable
Vibe Code Bench
Grok 4.3—Kimi K2.637.89%SourceNot directly comparable
cursorBench31
Grok 4.3—Kimi K2.647.6%SourceNot directly comparable
Knowledge
GPQA Diamond (Vals)
Grok 4.391.4%SourceKimi K2.689.1%SourceGrok 4.3 leads this result
MMLU-Pro (Vals)
Grok 4.385.8%SourceKimi K2.687.6%SourceKimi K2.6 leads this result
GPQA
Grok 4.3—Kimi K2.690.5%SourceNot directly comparable
GPQA-D
Grok 4.3—Kimi K2.690.5%SourceNot directly comparable
HLE
Grok 4.3—Kimi K2.634.7%SourceNot directly comparable
Math
AIME26
Grok 4.3—Kimi K2.696.4%SourceNot directly comparable
HMMT Feb 2026
Grok 4.3—Kimi K2.692.7%SourceNot directly comparable
MMAnswerBench
Grok 4.3—Kimi K2.686.0%SourceNot directly comparable
FrontierMath v2 (Tiers 1-3)
Grok 4.3—Kimi K2.638.966%SourceNot directly comparable
FrontierMath v2 (Tier 4)
Grok 4.3—Kimi K2.614.580%SourceNot directly comparable
Multimodal
MMMU-Pro
Grok 4.3—Kimi K2.679.4%SourceNot directly comparable
MMMU-Pro w/ Python
Grok 4.3—Kimi K2.680.1%SourceNot directly comparable
CharXiv
Grok 4.3—Kimi K2.680.4%SourceNot directly comparable
MathVision
Grok 4.3—Kimi K2.687.4%SourceNot directly comparable
V*
Grok 4.3—Kimi K2.696.9%SourceNot directly comparable
Frequently asked questions
Which is better, Grok 4.3 or Kimi K2.6?
Kimi K2.6 has the higher public score estimate, 65.39 versus 59.73, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, Grok 4.3 or Kimi K2.6?
Kimi K2.6 leads the public coding lane, 50.8 to 33.6, with Supported evidence for both models, although the 90% intervals overlap.
Which is better for agentic tasks, Grok 4.3 or Kimi K2.6?
Kimi K2.6 leads the public agentic tasks lane, 46.7 to 26.9, with Supported evidence for both models, although the 90% intervals overlap.
Which costs less, Grok 4.3 or Kimi K2.6?
For the stated presets, chat costs $0.0025 on Grok 4.3 and $0.00295 on Kimi K2.6; repository review costs $0.07 and $0.0595; the cache-heavy agent loop costs $0.09 and $0.249. Kimi K2.6 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Grok 4.3 or Kimi K2.6?
Grok 4.3 has the larger documented context window: 1M, compared with 256K.
Related comparisons
Last updated September 10, 2026
Watch Grok 4.3 vs Kimi K2.6
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.