Knowledge
Directional only- Kimi K2.5
- 56.9
- ZAYA1-8B
- 73.6
- Weighted basis
- 4 vs 2 rows
- Reading
- Directional only
Model comparison
Kimi K2.5 vs ZAYA1-8B
Updated August 5, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Kimi K2.5 has the higher public score, 58.83 versus 31.58, and the 90% score intervals do not overlap.
7 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.
Long documents
Prompts that approach the documented context limit
Kimi K2.5
Kimi K2.5 has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Chat 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-rates
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. ZAYA1-8B 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. ZAYA1-8B has no comparable published API token rate.
Confidence: rate-fallback
Repository 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
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
38 Kimi K2.5 only
7 Shared
4 ZAYA1-8B only
- Shared results
- 7
- Kimi K2.5 only
- 38
- ZAYA1-8B only
- 4
- Like-for-like categories
- 0 / 8
3 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.
Math
Directional only- Kimi K2.5
- 60.6
- ZAYA1-8B
- 80.4
- Weighted basis
- 4 vs 2 rows
- Reading
- Directional only
Instruction following
Directional only- Kimi K2.5
- 93.9
- ZAYA1-8B
- 64.1
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Agentic
Not comparable- Kimi K2.5
- 55.0
- ZAYA1-8B
- Not measured
- Weighted basis
- 2 vs 0 rows
- Reading
- Not comparable
Coding
Not comparable- Kimi K2.5
- 59.4
- ZAYA1-8B
- Not measured
- Weighted basis
- 4 vs 0 rows
- Reading
- Not comparable
Reasoning
Not comparable- Kimi K2.5
- 61.0
- ZAYA1-8B
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K2.5
- 82.3
- ZAYA1-8B
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Multimodal
Not comparable- Kimi K2.5
- 78.5
- ZAYA1-8B
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
| Category | Kimi K2.5 | ZAYA1-8B | Weighted basis | Reading |
|---|---|---|---|---|
| Knowledge | 56.9 | 73.6 | Directional only4 vs 2 rows | Directional only |
| Math | 60.6 | 80.4 | Directional only4 vs 2 rows | Directional only |
| Instruction following | 93.9 | 64.1 | Directional only1 vs 2 rows | Directional only |
| Agentic | 55.0 | Not measured | Not comparable2 vs 0 rows | Not comparable |
| Coding | 59.4 | Not measured | Not comparable4 vs 0 rows | Not comparable |
| Reasoning | 61.0 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Multilingual | 82.3 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Multimodal | 78.5 | Not measured | Not comparable1 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.
GPQA
Knowledge
HMMT Feb 2026
Math
MMLU-Pro
Knowledge
IFEval
Instruction following
AIME26
Math
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
- ZAYA1-8B
- Self-hosted; infrastructure cost varies
- Fits in one request
ZAYA1-8B has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Kimi K2.5
- $0.039
- Fits in one request
- ZAYA1-8B
- Self-hosted; infrastructure cost varies
- Fits in one request
ZAYA1-8B 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
- ZAYA1-8B
- Self-hosted; infrastructure cost varies
- Does not fit in one request
- Cached-input rate unavailable
ZAYA1-8B 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. ZAYA1-8B has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationKimi K2.5ZAYA1-8B
Context window
Maximum documented context; output-token limits may be lower.
Kimi K2.5
256K
ZAYA1-8B
131K
API model ID
Kimi K2.5
Not sourced
ZAYA1-8B
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
ZAYA1-8B
No comparable hosted API rate
Documented inputs
Kimi K2.5
Not sourced
ZAYA1-8B
Not sourced
Documented outputs
Kimi K2.5
Not sourced
ZAYA1-8B
Not sourced
Provider availability
Kimi K2.5
Not sourced
ZAYA1-8B
Not sourced
Reasoning profile
Kimi K2.5
Non-Reasoning
ZAYA1-8B
Reasoning
Weight access
Kimi K2.5
Open Weight
ZAYA1-8B
Open Weight
License
Kimi K2.5
Open Weight
ZAYA1-8B
Open Weight
Release date
Kimi K2.5
2026-02-01
ZAYA1-8B
2026-05-05
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, 58.83 versus 31.58, and the 90% score intervals do not overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- 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
ZAYA1-8B
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 evidence49 rows
Agentic
Terminal-Bench 2.0
Kimi K2.550.8%SourceZAYA1-8B—Not directly comparable
BrowseComp
Kimi K2.560.6%SourceZAYA1-8B—Not directly comparable
Claw-Eval
Kimi K2.552.3%SourceZAYA1-8B—Not directly comparable
QwenClawBench
Kimi K2.554.3%SourceZAYA1-8B—Not directly comparable
τ³-bench results
Kimi K2.565.7%SourceZAYA1-8B—Not directly comparable
DeepSearchQA
Kimi K2.577.1%SourceZAYA1-8B—Not directly comparable
DeepPlanning
Kimi K2.514.4%SourceZAYA1-8B—Not directly comparable
Toolathlon
Kimi K2.527.8%SourceZAYA1-8B—Not directly comparable
MCP Atlas
Kimi K2.529.5%SourceZAYA1-8B—Not directly comparable
MCP-Tasks
Kimi K2.559.1%SourceZAYA1-8B—Not directly comparable
WideResearch
Kimi K2.572.7%SourceZAYA1-8B—Not directly comparable
Gert Labs
Kimi K2.545.88%SourceZAYA1-8B—Not directly comparable
ResearchClawBench
Kimi K2.514.0%SourceZAYA1-8B—Not directly comparable
JobBench
Kimi K2.58.7%SourceZAYA1-8B—Not directly comparable
BFCL v4
Kimi K2.5—ZAYA1-8B39.2%SourceNot directly comparable
Coding
SWE-bench Verified
Kimi K2.576.8%SourceZAYA1-8B—Not directly comparable
SWE-bench Verified*
Kimi K2.570.8%SourceZAYA1-8B—Not directly comparable
LiveCodeBench v6
Kimi K2.585.0%SourceZAYA1-8B65.8%SourceKimi K2.5 leads this result
SWE-bench Pro
Kimi K2.550.7%SourceZAYA1-8B—Not directly comparable
SWE Multilingual
Kimi K2.573%SourceZAYA1-8B—Not directly comparable
SWE-Rebench
Kimi K2.558.5%SourceZAYA1-8B—Not directly comparable
React Native Evals
Kimi K2.577.2%SourceZAYA1-8B—Not directly comparable
SciCode
Kimi K2.548.7%SourceZAYA1-8B—Not directly comparable
Reasoning
LongBench v2
Kimi K2.561%SourceZAYA1-8B—Not directly comparable
Knowledge
GPQA
Kimi K2.587.6%SourceZAYA1-8B71%SourceKimi K2.5 leads this result
GPQA-D
Kimi K2.587.6%SourceZAYA1-8B71.0%SourceKimi K2.5 leads this result
SuperGPQA
Kimi K2.569.2%SourceZAYA1-8B—Not directly comparable
MMLU-Pro
Kimi K2.587.1%SourceZAYA1-8B74.2%SourceKimi K2.5 leads this result
MMLU-Pro (Arcee)
Kimi K2.587.1%SourceZAYA1-8B—Not directly comparable
HLE
Kimi K2.530.1%SourceZAYA1-8B—Not directly comparable
Math
AIME 2025
Kimi K2.596.1%SourceZAYA1-8B—Not directly comparable
AIME26
Kimi K2.595.8%SourceZAYA1-8B89.1%SourceKimi K2.5 leads this result
AIME25 (Arcee)
Kimi K2.596.3%SourceZAYA1-8B—Not directly comparable
HMMT Feb 2025
Kimi K2.595.4%SourceZAYA1-8B—Not directly comparable
HMMT Nov 2025
Kimi K2.591.1%SourceZAYA1-8B—Not directly comparable
HMMT Feb 2026
Kimi K2.587.1%SourceZAYA1-8B71.6%SourceKimi K2.5 leads this result
MMAnswerBench
Kimi K2.581.8%SourceZAYA1-8B—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Kimi K2.527.900%SourceZAYA1-8B—Not directly comparable
FrontierMath v2 (Tier 4)
Kimi K2.54.200%SourceZAYA1-8B—Not directly comparable
IMOAnswerBench
Kimi K2.5—ZAYA1-8B59.3%SourceNot directly comparable
Apex
Kimi K2.5—ZAYA1-8B32.2%SourceNot directly comparable
Multilingual
Multimodal
Instruction following
Frequently asked questions
Which is better, Kimi K2.5 or ZAYA1-8B?
Kimi K2.5 has the higher public score, 58.83 versus 31.58, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.
Which is better for coding, Kimi K2.5 or ZAYA1-8B?
The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.
Which is better for agentic tasks, Kimi K2.5 or ZAYA1-8B?
The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.
Which costs less, Kimi K2.5 or ZAYA1-8B?
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 ZAYA1-8B?
Kimi K2.5 has the larger documented context window: 256K, compared with 131K.
Related comparisons
Last updated August 5, 2026
Watch Kimi K2.5 vs ZAYA1-8B
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.