Kimi K2.6 model profile
Evidence snapshot:
Kimi K2.6 is a Moonshot AI model profile with a 262,144-token context window. The dated catalog snapshot records text, image modalities and interfaces for prompt caching, reasoning controls, structured output, text generation, tool calling.
Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and can convert prompts and visual inputs into production-ready interfaces. Its agent swarm architecture scales to hundreds of parallel sub-agents for autonomous task decomposition - delivering documents, websites, and spreadsheets in a single run without human oversight.
Kimi K2.6 is a reference profile: Moonshot AI publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call.
Kimi K2.6 is an independent knowledge profile. It does not claim that Kendr hosts the model, quote a Kendr customer price, promise API availability, or advertise a Kendr routing receipt.
Kimi K2.6 accepts text and image and returns text. The snapshot records prompt caching, reasoning controls, structured output, text generation, tool calling, and vision as capabilities or interfaces.
The dated reference snapshot lists $0.53675 input and $2.26 output per million tokens. Cached input is $0.058 per million tokens.
The 2026-09-06 snapshot includes 2 Artificial Analysis indexes, 21 Design Arena categories, and AutoExacto results for 20 provider observations. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable.
Kimi K2.6 ranked #53 in the cited trailing-7-day third-party catalog usage dataset as of 2026-09-06. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.
Profile data was reviewed for the 2026-09-06 snapshot. Claims retain their source dates, and unavailable fields are shown as unavailable instead of estimated.
Profile facts
- Profile type
- Model research profile
- Provider or publisher
- Moonshot AI
- Context window
- 262,144 tokens
- Snapshot date
- 2026-09-06
- Knowledge cutoff
- Not disclosed
- Reference catalog ID
- moonshotai/kimi-k2.6
When to pick Kimi K2.6
Kimi K2.6 is a reference profile: Moonshot AI publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call.
Each condition below is derived from this model's own values in the 2026-09-06 snapshot, compared only against models publishing the same field.
- Reach for it when the input includes images. The snapshot records image input, so screenshots, scans, and diagrams can be sent directly instead of described.
- Reach for it when the model has to call tools. Tool calling is recorded in the snapshot, so this route can drive an agent loop rather than only answer in prose.
- Look elsewhere when you need to call it through Kendr today. This is a research reference profile with no Kendr alias. The facts and evidence here are published for comparison; the model is not routable on Kendr.
Context, modalities, and identity
Kimi K2.6 accepts text and image and returns text. The snapshot records prompt caching, reasoning controls, structured output, text generation, tool calling, and vision as capabilities or interfaces.
- Provider or publisher
- Moonshot AI
- Context window
- 262,144 tokens
- Maximum output
- 235,929 tokens
- Input modalities
- text and image
- Output modalities
- text
- Knowledge cutoff
- Not disclosed
Model overview
Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and can convert prompts and visual inputs into production-ready interfaces. Its agent swarm architecture scales to hundreds of parallel sub-agents for autonomous task decomposition - delivering documents, websites, and spreadsheets in a single run without human oversight.
- Reference model ID
- moonshotai/kimi-k2.6
- Canonical version
- moonshotai/kimi-k2.6-20260420
- Hugging Face ID
- moonshotai/Kimi-K2.6
Capabilities and supported controls
Capabilities and parameters are reported from the dated catalog and source records; their presence does not guarantee identical behavior across every provider route.
- Supported parameters
- frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, parallel_tool_calls, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p
- prompt caching
- reasoning controls
- structured output
- text generation
- tool calling
- vision
Dated pricing snapshot
The dated reference snapshot lists $0.53675 input and $2.26 output per million tokens. Cached input is $0.058 per million tokens.
These are dated third-party reference prices, not Kendr prices or an availability offer. Routes, tiers, caching, region, and provider policy can change the landed price.
- Price date
- 2026-09-06
- Input
- $0.53675 per 1M tokens
- Cached input
- $0.058 per 1M tokens
- Output
- $2.26 per 1M tokens
Provider routes, performance, and uptime
The snapshot retains 21 provider endpoints. Provider prices, context limits, p50 performance, and uptime can differ by route and are not Kendr guarantees.
- Active provider endpoints
- 18
- Best p50 latency
- 0.35 s
- Best p50 throughput
- 237 tok/s
- Availability with routing
- 99.96% over the sampled window
- Availability without routing
- 94.81% over the sampled window
- Performance date
- 2026-09-06
| Provider | Quantization | Input / 1M | Output / 1M | Cache read / 1M | Context | p50 latency | p50 throughput | Uptime (1d) |
|---|---|---|---|---|---|---|---|---|
| Baidu Qianfan | fp4 | $0.53675 / 1M | $2.26 / 1M | $0.0904 / 1M | 262,144 tokens | 1.55 s | 34 tok/s | 99.98% |
| Decart | fp4 | $0.5372 / 1M | $2.2618 / 1M | $0.0905 / 1M | 262,144 tokens | 0.39 s | 85 tok/s | 99.64% |
| Inceptron | int4 | $0.56 / 1M | $3.39 / 1M | $0.17 / 1M | 262,144 tokens | 0.52 s | 66 tok/s | 99.80% |
| Chutes | int4 | $0.58 / 1M | $3.4 / 1M | $0.058 / 1M | 262,144 tokens | 2.37 s | 37 tok/s | 99.43% |
| StreamLake | fp8 | $0.5985 / 1M | $2.52 / 1M | $0.1008 / 1M | 256,000 tokens | 1.31 s | 60 tok/s | 99.46% |
| CoreWeave | fp4 | $0.65 / 1M | $3.41 / 1M | $0.15 / 1M | 262,144 tokens | 0.35 s | 237 tok/s | 99.98% |
| Crusoe | bf16 | $0.7 / 1M | $3.5 / 1M | $0.35 / 1M | 262,144 tokens | 0.50 s | 40 tok/s | 99.27% |
| DeepInfra | fp4 | $0.75 / 1M | $3.5 / 1M | $0.15 / 1M | 262,144 tokens | 1.43 s | 29 tok/s | 99.60% |
| Venice | int4 | $0.75 / 1M | $3.5 / 1M | $0.16 / 1M | 256,000 tokens | 1.27 s | 27 tok/s | 99.66% |
| Parasail | int4 | $0.75 / 1M | $3.5 / 1M | $0.16 / 1M | 262,144 tokens | 0.88 s | 107 tok/s | 99.92% |
| SiliconFlow | fp8 | $0.77 / 1M | $3.4 / 1M | $0.14 / 1M | 262,144 tokens | 1.42 s | 33 tok/s | 99.26% |
| NovitaAI | unknown | $0.8 / 1M | $3.4 / 1M | $0.16 / 1M | 262,144 tokens | 1.32 s | 31 tok/s | 99.69% |
| GMICloud | fp8 | $0.855 / 1M | $3.6 / 1M | $0.144 / 1M | 262,144 tokens | 3.44 s | 147 tok/s | 98.93% |
| DigitalOcean | unknown | $0.95 / 1M | $4 / 1M | $0.19 / 1M | 262,144 tokens | 1.11 s | 45 tok/s | 99.68% |
| AtlasCloud | int4 | $0.95 / 1M | $4 / 1M | $0.16 / 1M | 262,144 tokens | 1.78 s | 34 tok/s | 99.42% |
| Cloudflare | unknown | $0.95 / 1M | $4 / 1M | $0.16 / 1M | 262,144 tokens | 0.65 s | 54 tok/s | 99.96% |
| Moonshot AI | int4 | $0.95 / 1M | $4 / 1M | $0.16 / 1M | 262,144 tokens | 0.86 s | 61 tok/s | 99.97% |
| Phala | unknown | $1.09 / 1M | $4.6 / 1M | $0.37 / 1M | 262,144 tokens | No recent p50 | No recent p50 | 0.00% |
| Baseten (US) | fp4 | $0.95 / 1M | $4 / 1M | $0.16 / 1M | 262,000 tokens | 0.58 s | 144 tok/s | 44.27% |
| Fireworks | unknown | $0.95 / 1M | $4 / 1M | $0.16 / 1M | 262,144 tokens | 0.86 s | 49.5 tok/s | 0.00% |
| Baseten | fp4 | $0.95 / 1M | $4 / 1M | $0.16 / 1M | 262,000 tokens | 0.47 s | 27 tok/s | No recent uptime |
Benchmark evidence and limitations
The 2026-09-06 snapshot includes 2 Artificial Analysis indexes, 21 Design Arena categories, and AutoExacto results for 20 provider observations. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable.
Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.
- Coding Index
- 61.8
- Agentic Index
- 22.2
- AutoExacto coverage
- 20 provider observations over 32 days
| Design Arena category | Elo | Win rate | Rank |
|---|---|---|---|
| agenticgamedev | 1140 | 47.4% | #19 |
| agentichtmlslides | 1248 | 59% | #2 |
| agenticslides | 1187 | 45.8% | #4 |
| agenticslides(html) | 1252 | 59.2% | #2 |
| agenticslides(python-pptx) | 1186 | 45.5% | #4 |
| androidnative | 1180 | 50.2% | #20 |
| fullstack | 1161 | 51.8% | #22 |
| godotgamedev | 1160 | 47.1% | #14 |
| htmlslides | 1192 | 52.3% | #9 |
| mobileapps | 1189 | 49% | #21 |
| pptxslides | 1181 | 44.3% | #4 |
| python-pptxslides | 1180 | 42.1% | #14 |
| webapps | 1268 | 59.3% | #7 |
| 3d | 1306 | 57.6% | #16 |
| asciiart | 1190 | 47.5% | #25 |
| codecategories | 1289 | 54.8% | #19 |
| dataviz | 1271 | 52.1% | #22 |
| gamedev | 1281 | 54.9% | #26 |
| svg | 1217 | 51.2% | #26 |
| uicomponent | 1288 | 55.6% | #22 |
| website | 1283 | 54.2% | #20 |
AutoExacto provider benchmarks
Rolling provider observations from the cited third-party benchmark view. The lookback is 32 days; GPQA Diamond and TAU-Bench Airline measure different abilities and should not be blended into one score.
| Provider | GPQA Diamond | TAU-Bench Airline | Runs |
|---|---|---|---|
| AtlasCloud | 79.34% | 65.94% | 4 |
| auto-routing | 82.62% | 73.57% | 4 |
| Baidu Qianfan | 85.39% | 72.53% | 4 |
| Chutes | 85.13% | 69.44% | 4 |
| Cloudflare | 84.15% | 74.19% | 4 |
| CoreWeave | 85.71% | 73.99% | 4 |
| Crusoe | 79.19% | 71.61% | 4 |
| Decart | 86.45% | 65.76% | 4 |
| DeepInfra | 55.56% | 75.48% | 4 |
| DigitalOcean | 75.73% | 72.69% | 4 |
| GMICloud | — | 71.33% | 1 |
| Inceptron | 86.92% | 70.81% | 4 |
| Moonshot AI | 86.64% | 76.08% | 4 |
| NovitaAI | 85.87% | 75.13% | 4 |
| Parasail | 86.51% | 70.95% | 4 |
| Phala | 82.63% | 74.39% | 4 |
| SiliconFlow | 88.26% | 74.06% | 4 |
| StreamLake | 84.8% | 72.95% | 4 |
| Together | 85.56% | — | 2 |
| Venice | 59.16% | 72.89% | 4 |
Popularity and market context
Kimi K2.6 ranked #53 in the cited trailing-7-day third-party catalog usage dataset as of 2026-09-06. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.
- Third-party catalog rank
- #53
- Global market share
- Not inferred
- Observation date
- 2026-09-06
Frequently asked questions
When should I use Kimi K2.6?
Kimi K2.6 is a reference profile: Moonshot AI publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call. Reach for it when the input includes images: The snapshot records image input, so screenshots, scans, and diagrams can be sent directly instead of described. Reach for it when the model has to call tools: Tool calling is recorded in the snapshot, so this route can drive an agent loop rather than only answer in prose. Look elsewhere when you need to call it through Kendr today: This is a research reference profile with no Kendr alias. The facts and evidence here are published for comparison; the model is not routable on Kendr.
What is Kimi K2.6?
Kimi K2.6 is a Moonshot AI model profile with a 262,144-token context window. The dated catalog snapshot records text, image modalities and interfaces for prompt caching, reasoning controls, structured output, text generation, tool calling.
What context window does Kimi K2.6 have?
The dated profile lists 262,144 tokens of context and up to 235,929 output tokens. Provider routes, variants, and runtime configuration can impose lower effective limits.
What benchmark evidence is available for Kimi K2.6?
The 2026-09-06 snapshot includes 2 Artificial Analysis indexes, 21 Design Arena categories, and AutoExacto results for 20 provider observations. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable. Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.
Is Kimi K2.6 available through Kendr?
This is an independent knowledge profile, not a Kendr-hosted availability claim. Check Kendr's live public model API for currently enabled Kendr aliases.
Does the popularity rank represent Kimi K2.6's global market share?
Kimi K2.6 ranked #53 in the cited trailing-7-day third-party catalog usage dataset as of 2026-09-06. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share. No global market-share percentage is inferred when the source does not publish one.
Sources and evidence dates
- Kimi K2.6 third-party catalog record (catalog, checked 2026-09-06)
- Third-party model catalog methodology (methodology, checked 2026-09-06)
- Moonshot AI official model documentation (primary, checked 2026-09-06)
- Artificial Analysis capability indices methodology (benchmark, checked 2026-09-06)
- Design Arena leaderboard and methodology (benchmark, checked 2026-09-06)
- Kimi K2.6 third-party catalog record (benchmark, checked 2026-09-06)