Agentic
Not comparable- Kimi K2.6
- 46.1
- Estimated · #88/151
- Trinity-Large-Preview
- Not ranked
- Basis
- BenchAlign lane · 12 vs 0 public rows
- Reading
- Not comparable
Kimi K2.6 vs Trinity-Large-Preview
Updated September 4, 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, 65.3 versus 55.39, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
1 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
Trinity-Large-Preview
Trinity-Large-Preview has the larger documented context window.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
Trinity-Large-Preview
Trinity-Large-Preview 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
Trinity-Large-Preview
Trinity-Large-Preview 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. Trinity-Large-Preview has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
Show secondary and unsupported calls
Repository review cost
50K fresh input + 3K output tokens
Trinity-Large-Preview
Trinity-Large-Preview 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
Trinity-Large-Preview is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Trinity-Large-Preview is not ranked on the public lane for agentic, so no winner is named for agentic.
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.
36 Kimi K2.6 only
1 Shared
3 Trinity-Large-Preview only
- Shared results
- 1
- Kimi K2.6 only
- 36
- Trinity-Large-Preview only
- 3
- Like-for-like categories
- 0 / 8
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
Not comparable- Kimi K2.6
- 51.6
- Supported · #62/183
- Trinity-Large-Preview
- Not ranked
- Basis
- BenchAlign lane · 10 vs 0 public rows
- Reading
- Not comparable
Reasoning
Not comparable- Kimi K2.6
- Not ranked
- Trinity-Large-Preview
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Kimi K2.6
- 61.7
- Supported · #34/181
- Trinity-Large-Preview
- Not ranked
- Basis
- BenchAlign lane · 5 vs 3 public rows
- Reading
- Not comparable
Math
Not comparable- Kimi K2.6
- 71.3
- #1/7
- Trinity-Large-Preview
- Not ranked
- Basis
- Provisional lane · 4 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K2.6
- Not ranked
- Trinity-Large-Preview
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Kimi K2.6
- 63.2
- #26/48
- Trinity-Large-Preview
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Kimi K2.6
- Not ranked
- Trinity-Large-Preview
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Kimi K2.6 | Trinity-Large-Preview | Basis | Reading |
|---|---|---|---|---|
| Agentic | 46.1Estimated · #88/151 | Not ranked | Not comparableBenchAlign lane · 12 vs 0 public rows | Not comparable |
| Coding | 51.6Supported · #62/183 | Not ranked | Not comparableBenchAlign lane · 10 vs 0 public rows | Not comparable |
| Reasoning | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Knowledge | 61.7Supported · #34/181 | Not ranked | Not comparableBenchAlign lane · 5 vs 3 public rows | Not comparable |
| Math | 71.3#1/7 | Not ranked | Not comparableProvisional lane · 4 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | 63.2#26/48 | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | 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.
A shared-evidence shape is not available.
BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.
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
- Trinity-Large-Preview
- $0.00075
- Fits in one request
Trinity-Large-Preview 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
- Trinity-Large-Preview
- $0.0155
- Fits in one request
Trinity-Large-Preview 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
- Trinity-Large-Preview
- $0.065
- Fits in one request
- Cached input priced at the published list-input rate
Trinity-Large-Preview has the lower modeled cost
Kimi K2.6 has no published cached-input rate, so cached tokens use its listed input rate. Trinity-Large-Preview 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 K2.6Trinity-Large-Preview
Context window
Maximum documented context; output-token limits may be lower.
Kimi K2.6
256K
Trinity-Large-Preview
512K
API model ID
Kimi K2.6
Not sourced
Trinity-Large-Preview
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
Trinity-Large-Preview
Not published
Documented inputs
Kimi K2.6
Not sourced
Trinity-Large-Preview
Not sourced
Documented outputs
Kimi K2.6
Not sourced
Trinity-Large-Preview
Not sourced
Provider availability
Kimi K2.6
Not sourced
Trinity-Large-Preview
Not sourced
Reasoning profile
Kimi K2.6
Reasoning
Trinity-Large-Preview
Non-Reasoning
Weight access
Kimi K2.6
Open Weight
Trinity-Large-Preview
Open Weight
License
Kimi K2.6
Open Weight
Trinity-Large-Preview
Open Weight
Release date
Kimi K2.6
2026-04-20
Trinity-Large-Preview
2026-01-27
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, 65.3 versus 55.39, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.0595 vs $0.0155. Cache-heavy agent loop: $0.249 vs $0.065.
- Context tradeoff
- Trinity-Large-Preview has the larger documented window (512K).
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
Trinity-Large-Preview
API / mo$938
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 evidence40 rows
Agentic
Terminal-Bench 2.0
Kimi K2.666.7%SourceTrinity-Large-Preview—Not directly comparable
BrowseComp
Kimi K2.683.2%SourceTrinity-Large-Preview—Not directly comparable
OSWorld-Verified
Kimi K2.673.1%SourceTrinity-Large-Preview—Not directly comparable
Toolathlon
Kimi K2.650%SourceTrinity-Large-Preview—Not directly comparable
MCP Atlas
Kimi K2.655.9%SourceTrinity-Large-Preview—Not directly comparable
Claw-Eval
Kimi K2.662.3%SourceTrinity-Large-Preview—Not directly comparable
DeepSearchQA
Kimi K2.692.5%SourceTrinity-Large-Preview—Not directly comparable
WideResearch
Kimi K2.680.8%SourceTrinity-Large-Preview—Not directly comparable
Gert Labs
Kimi K2.656.82%SourceTrinity-Large-Preview—Not directly comparable
ResearchClawBench
Kimi K2.618.0%SourceTrinity-Large-Preview—Not directly comparable
OSWorld 2.0
Kimi K2.64.6%SourceTrinity-Large-Preview—Not directly comparable
Terminal-Bench 2.1 (Vals)
Kimi K2.653.6%SourceTrinity-Large-Preview—Not directly comparable
Coding
SWE-bench Verified
Kimi K2.680.2%SourceTrinity-Large-Preview—Not directly comparable
LiveCodeBench v6
Kimi K2.689.6%SourceTrinity-Large-Preview—Not directly comparable
SWE-bench Pro
Kimi K2.658.6%SourceTrinity-Large-Preview—Not directly comparable
SWE Multilingual
Kimi K2.676.7%SourceTrinity-Large-Preview—Not directly comparable
SciCode
Kimi K2.652.2%SourceTrinity-Large-Preview—Not directly comparable
Terminal-Bench 2.0
Kimi K2.666.7%SourceTrinity-Large-Preview—Not directly comparable
Vibe Code Bench
Kimi K2.637.89%SourceTrinity-Large-Preview—Not directly comparable
cursorBench31
Kimi K2.647.6%SourceTrinity-Large-Preview—Not directly comparable
LiveCodeBench (Vals)
Kimi K2.686.8%SourceTrinity-Large-Preview—Not directly comparable
SWE-bench (Vals)
Kimi K2.676.2%SourceTrinity-Large-Preview—Not directly comparable
Knowledge
GPQA
Kimi K2.690.5%SourceTrinity-Large-Preview—Not directly comparable
GPQA-D
Kimi K2.690.5%SourceTrinity-Large-Preview63.3%SourceKimi K2.6 leads this result
HLE
Kimi K2.634.7%SourceTrinity-Large-Preview—Not directly comparable
GPQA Diamond (Vals)
Kimi K2.689.1%SourceTrinity-Large-Preview—Not directly comparable
MMLU-Pro (Vals)
Kimi K2.687.6%SourceTrinity-Large-Preview—Not directly comparable
MMLU
Kimi K2.6—Trinity-Large-Preview87.2%SourceNot directly comparable
MMLU-Pro (Arcee)
Kimi K2.6—Trinity-Large-Preview75.2%SourceNot directly comparable
Math
AIME26
Kimi K2.696.4%SourceTrinity-Large-Preview—Not directly comparable
HMMT Feb 2026
Kimi K2.692.7%SourceTrinity-Large-Preview—Not directly comparable
MMAnswerBench
Kimi K2.686.0%SourceTrinity-Large-Preview—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Kimi K2.638.966%SourceTrinity-Large-Preview—Not directly comparable
FrontierMath v2 (Tier 4)
Kimi K2.614.580%SourceTrinity-Large-Preview—Not directly comparable
AIME25 (Arcee)
Kimi K2.6—Trinity-Large-Preview24.0%SourceNot directly comparable
Multimodal
MMMU-Pro
Kimi K2.679.4%SourceTrinity-Large-Preview—Not directly comparable
MMMU-Pro w/ Python
Kimi K2.680.1%SourceTrinity-Large-Preview—Not directly comparable
CharXiv
Kimi K2.680.4%SourceTrinity-Large-Preview—Not directly comparable
MathVision
Kimi K2.687.4%SourceTrinity-Large-Preview—Not directly comparable
V*
Kimi K2.696.9%SourceTrinity-Large-Preview—Not directly comparable
Frequently asked questions
Which is better, Kimi K2.6 or Trinity-Large-Preview?
Kimi K2.6 has the higher public score estimate, 65.3 versus 55.39, 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 Trinity-Large-Preview?
Trinity-Large-Preview is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, Kimi K2.6 or Trinity-Large-Preview?
Trinity-Large-Preview is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, Kimi K2.6 or Trinity-Large-Preview?
For the stated presets, chat costs $0.00295 on Kimi K2.6 and $0.00075 on Trinity-Large-Preview; repository review costs $0.0595 and $0.0155; the cache-heavy agent loop costs $0.249 and $0.065. Kimi K2.6 has no published cached-input rate, so cached tokens use its listed input rate. Trinity-Large-Preview has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Kimi K2.6 or Trinity-Large-Preview?
Trinity-Large-Preview has the larger documented context window: 512K, compared with 256K.
Related comparisons
Last updated September 4, 2026
Watch Kimi K2.6 vs Trinity-Large-Preview
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.