Long documents
Prompts that approach the documented context limit
Sakana Fugu
Sakana Fugu has the larger documented context window.
Confidence: documented
Kimi K2.5 vs Sakana Fugu
Updated September 27, 2026. We do not rank this pair: at least one has no public score. Public scores include evidence status and uncertainty.
Both of these models will change. Get the price, version and retirement notices for the pair, each with its source. Follow model changes
Share or export
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. 5 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 5
- Kimi K2.5 only
- 40
- Sakana Fugu only
- 6
- Like-for-like categories
- 0 / 8
Supported: Kimi K2.5How the comparison works
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.
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Sakana Fugu is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limitedAgentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Sakana Fugu is not ranked on the public lane for agentic, so no winner is named for agentic.
Confidence: limitedChat 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-ratesCache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: rate-fallbackRepository 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
Which one for a specific job
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
38.9Kimi K2.5—Sakana Fugu
Not comparable · BenchAlign v5.7
The coding row is not comparable on the public lane: at least one model is not measured or not ranked there.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
40 Kimi K2.5 only
5 Shared
6 Sakana Fugu only
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.
SciCodeCoding
Normalized gap 11.4Kimi K2.5:48.7%Sakana Fugu:60.1%SWE-bench ProCoding
Normalized gap 8.3Kimi K2.5:50.7%Sakana Fugu:59%GPQAKnowledge
Normalized gap 7.9Kimi K2.5:87.6%Sakana Fugu:95.5%
Bars run 0–100 on each benchmark’s normalized display scale
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign v5.7 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.
Agentic
Not comparable- Kimi K2.5
- 36.1
- Estimated · #52/105
- Sakana Fugu
- Not ranked
- Basis
- BenchAlign v5.7 lane · 14 vs 1 public rows
- Reading
- Not comparable
Coding
Not comparable- Kimi K2.5
- 38.9
- Estimated · #61/135
- Sakana Fugu
- Not ranked
- Basis
- BenchAlign v5.7 lane · 8 vs 5 public rows
- Reading
- Not comparable
Reasoning
Not comparable- Kimi K2.5
- 55.1
- Unranked · 3 rankable rows
- Sakana Fugu
- 70.6
- Unranked · 1 rankable row
- Basis
- Provisional lane · 1 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Kimi K2.5
- 65.8
- #25/50
- Sakana Fugu
- 68.6
- Unranked · 1 rankable row
- Basis
- Provisional lane · 1 vs 1 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Kimi K2.5
- 47.8
- Estimated · #65/158
- Sakana Fugu
- Not ranked
- Basis
- BenchAlign v5.7 lane · 6 vs 3 public rows
- Reading
- Not comparable
Multilingual
Not comparable- Kimi K2.5
- 38.2
- #8/12
- Sakana Fugu
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Kimi K2.5
- 84.4
- #41/124
- Sakana Fugu
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Kimi K2.5
- 62.1
- #4/7
- Sakana Fugu
- Not ranked
- Basis
- Provisional lane · 4 vs 0 weighted rows
- Reading
- Not comparable
| Category | Kimi K2.5 | Sakana Fugu | Basis | Reading |
|---|---|---|---|---|
| Agentic | 36.1Estimated · #52/105 | Not ranked | Not comparableBenchAlign v5.7 lane · 14 vs 1 public rows | Not comparable |
| Coding | 38.9Estimated · #61/135 | Not ranked | Not comparableBenchAlign v5.7 lane · 8 vs 5 public rows | Not comparable |
| Reasoning | 55.1Unranked · 3 rankable rows | 70.6Unranked · 1 rankable row | Not comparableProvisional lane · 1 vs 1 weighted rows | Not comparable |
| Multimodal | 65.8#25/50 | 68.6Unranked · 1 rankable row | Not comparableProvisional lane · 1 vs 1 weighted rows | Not comparable |
| Knowledge | 47.8Estimated · #65/158 | Not ranked | Not comparableBenchAlign v5.7 lane · 6 vs 3 public rows | Not comparable |
| Multilingual | 38.2#8/12 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Instruction following | 84.4#41/124 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 62.1#4/7 | Not ranked | Not comparableProvisional lane · 4 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.7) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Bars run 0–100Methodology
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
- Sakana Fugu
- API rate not published
- Fits in one request
Sakana Fugu has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Kimi K2.5
- $0.039
- Fits in one request
- Sakana Fugu
- API rate not published
- Fits in one request
Sakana Fugu 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
- Sakana Fugu
- API rate not published
- Fits in one request
- Cached-input rate unavailable
Kimi K2.5 has no published cached-input rate, so cached tokens use its listed input rate. Sakana Fugu has no comparable published API token rate.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationKimi K2.5Sakana Fugu
Context window
Maximum documented context; output-token limits may be lower.
Kimi K2.5
256K
Sakana Fugu
1M
API model ID
Kimi K2.5
Not sourced
Sakana Fugu
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
Sakana Fugu
No comparable hosted API rate
Documented inputs
Kimi K2.5
Not sourced
Sakana Fugu
Not sourced
Documented outputs
Kimi K2.5
Not sourced
Sakana Fugu
Not sourced
Provider availability
Kimi K2.5
Not sourced
Sakana Fugu
Not sourced
Reasoning profile
Kimi K2.5
Non-Reasoning
Sakana Fugu
Reasoning
Weight access
Kimi K2.5
Open Weight
Sakana Fugu
Proprietary
License
Kimi K2.5
Open Weight
Sakana Fugu
Proprietary
Release date
Kimi K2.5
2026-02-01
Sakana Fugu
2026-06-22
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
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Sakana Fugu has the larger documented window (1M).
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Kimi K2.5 or Sakana Fugu?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, Kimi K2.5 or Sakana Fugu?
Sakana Fugu is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, Kimi K2.5 or Sakana Fugu?
Sakana Fugu is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, Kimi K2.5 or Sakana Fugu?
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 Sakana Fugu?
Sakana Fugu has the larger documented context window: 1M, compared with 256K.
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
Sakana Fugu
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 evidence51 rows
Agentic
Terminal-Bench 2.0
Kimi K2.550.8%SourceSakana Fugu—Not directly comparable
BrowseComp
Kimi K2.560.6%SourceSakana Fugu—Not directly comparable
Claw-Eval
Kimi K2.552.3%SourceSakana Fugu—Not directly comparable
QwenClawBench
Kimi K2.554.3%SourceSakana Fugu—Not directly comparable
τ³-bench results
Kimi K2.565.7%SourceSakana Fugu—Not directly comparable
DeepSearchQA
Kimi K2.577.1%SourceSakana Fugu—Not directly comparable
DeepPlanning
Kimi K2.514.4%SourceSakana Fugu—Not directly comparable
Toolathlon
Kimi K2.527.8%SourceSakana Fugu—Not directly comparable
MCP Atlas
Kimi K2.529.5%SourceSakana Fugu—Not directly comparable
MCP-Tasks
Kimi K2.559.1%SourceSakana Fugu—Not directly comparable
WideResearch
Kimi K2.572.7%SourceSakana Fugu—Not directly comparable
Gert Labs
Kimi K2.545.88%SourceSakana Fugu—Not directly comparable
ResearchClawBench
Kimi K2.514.0%SourceSakana Fugu—Not directly comparable
JobBench
Kimi K2.58.7%SourceSakana Fugu—Not directly comparable
Terminal-Bench 2.1
Kimi K2.5—Sakana Fugu80.2%SourceNot directly comparable
Coding
SWE-bench Verified
Kimi K2.576.8%SourceSakana Fugu—Not directly comparable
SWE-bench Verified*
Kimi K2.570.8%SourceSakana Fugu—Not directly comparable
LiveCodeBench v6
Kimi K2.585.0%SourceSakana Fugu92.9%SourceSakana Fugu leads this result
SWE-bench Pro
Kimi K2.550.7%SourceSakana Fugu59%SourceSakana Fugu leads this result
SWE Multilingual
Kimi K2.573%SourceSakana Fugu—Not directly comparable
SWE-Rebench
Kimi K2.558.5%SourceSakana Fugu—Not directly comparable
React Native Evals
Kimi K2.577.2%SourceSakana Fugu—Not directly comparable
SciCode
Kimi K2.548.7%SourceSakana Fugu60.1%SourceSakana Fugu leads this result
Terminal-Bench 2.1
Kimi K2.5—Sakana Fugu80.2%SourceNot directly comparable
LiveCodeBench Pro
Kimi K2.5—Sakana Fugu87.8%SourceNot directly comparable
Reasoning
Multimodal
MMMU-Pro
Kimi K2.578.5%SourceSakana Fugu—Not directly comparable
Video-MME
Kimi K2.587.4%SourceSakana Fugu—Not directly comparable
MMVU
Kimi K2.580.4%SourceSakana Fugu—Not directly comparable
VideoMMMU
Kimi K2.586.6%SourceSakana Fugu—Not directly comparable
CharXiv
Kimi K2.5—Sakana Fugu85.1%SourceNot directly comparable
Knowledge
GPQA
Kimi K2.587.6%SourceSakana Fugu95.5%SourceSakana Fugu leads this result
GPQA-D
Kimi K2.587.6%SourceSakana Fugu95.5%SourceSakana Fugu leads this result
SuperGPQA
Kimi K2.569.2%SourceSakana Fugu—Not directly comparable
MMLU-Pro
Kimi K2.587.1%SourceSakana Fugu—Not directly comparable
MMLU-Pro (Arcee)
Kimi K2.587.1%SourceSakana Fugu—Not directly comparable
HLE
Kimi K2.530.1%SourceSakana Fugu—Not directly comparable
HLE w/o tools
Kimi K2.5—Sakana Fugu47.2%SourceNot directly comparable
Multilingual
Instruction following
IFEval
Kimi K2.593.9%SourceSakana Fugu—Not directly comparable
Math
AIME 2025
Kimi K2.596.1%SourceSakana Fugu—Not directly comparable
AIME26
Kimi K2.595.8%SourceSakana Fugu—Not directly comparable
AIME25 (Arcee)
Kimi K2.596.3%SourceSakana Fugu—Not directly comparable
HMMT Feb 2025
Kimi K2.595.4%SourceSakana Fugu—Not directly comparable
HMMT Nov 2025
Kimi K2.591.1%SourceSakana Fugu—Not directly comparable
HMMT Feb 2026
Kimi K2.587.1%SourceSakana Fugu—Not directly comparable
MMAnswerBench
Kimi K2.581.8%SourceSakana Fugu—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Kimi K2.527.900%SourceSakana Fugu—Not directly comparable
FrontierMath v2 (Tier 4)
Kimi K2.54.200%SourceSakana Fugu—Not directly comparable
51 public results · 5 shared
Watch Kimi K2.5 vs Sakana Fugu
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.
Last updated September 27, 2026