Kimi K3 API model profile

Profile facts

Profile type

Kendr API model

Kendr API alias

kc-kimi-k3

Model developer

Moonshot AI

Kendr route provider

Moonshot Kimi

Context window

1.05M

Snapshot date

2026-09-06

Knowledge cutoff

Not disclosed

Reference catalog ID

moonshotai/kimi-k3

When to pick Kimi K3

Kimi K3 is worth reaching for when you need a route priced in the most expensive tenth of the catalog, scoring in the top quarter on the published intelligence index and carrying one of the widest context windows in the catalog. The conditions below are the ones its own numbers support.

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 a whole corpus has to fit in one prompt. The 1.05M-token context window is in the top quarter of the 163 models publishing a limit here, so long documents can go in whole rather than chunked.
• Reach for it when answer quality matters more than unit cost. It scores 50.2 on the Artificial Analysis intelligence index, above three quarters of the 32 models carrying that field in the 2026-09-06 snapshot.
• 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 the work does not justify the rate. At $2.55 in and $12.75 out per 1M tokens, this is in the most expensive tenth of the 146 models publishing both rates. Route routine or high-volume traffic to a cheaper model and keep this one for work that needs it.

Context, modalities, and identity

Kimi K3 accepts text, image, and video and returns text. The snapshot records agentic, coding, prompt caching, reasoning, reasoning controls, structured output, text generation, tool calling, tools, video input, vision, and web search as capabilities or interfaces.

Model developer

Moonshot AI

Kendr route provider

Moonshot Kimi

Context window

1.05M

Maximum output

943,718 tokens

Input modalities

text, image, and video

Output modalities

text

Knowledge cutoff

Not disclosed

Model overview

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.

Reference model ID

moonshotai/kimi-k3

Canonical version

moonshotai/kimi-k3-20260715

Hugging Face ID

moonshotai/Kimi-K3

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, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p

• agentic
• coding
• prompt caching
• reasoning
• reasoning controls
• structured output
• text generation
• tool calling
• tools
• video input
• vision
• web search

Dated pricing snapshot

The dated reference snapshot lists $2.55 input and $12.75 output per million tokens. Cached input is $0.256 per million tokens. The separately published provider-route reference is Route-specific (From $2.54 / $12.69 depending on the selected route; up to $3 / $15 ($0.30/M cached)).

Reference figures are dated 2026-09-06; the live Kendr quote can differ by selected provider route, context tier, caching, tools, region, and current rate card.

Price date

2026-09-06

Input

$2.55 per 1M tokens

Cached input

$0.256 per 1M tokens

Output

$12.75 per 1M tokens

Provider routes, performance, and uptime

The snapshot retains 18 provider endpoints. Provider prices, context limits, p50 performance, and uptime can differ by route and are not Kendr guarantees.

Active provider endpoints

17

Best p50 latency

0.89 s

Best p50 throughput

85 tok/s

Availability with routing

99.88% over the sampled window

Availability without routing

98.12% over the sampled window

Performance date

2026-09-06

Benchmark evidence and limitations

The 2026-09-06 snapshot includes 3 Artificial Analysis indexes, 15 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.

Intelligence Index

50.2

Coding Index

76.2

Agentic Index

50.9

AutoExacto coverage

20 provider observations over 32 days

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.

Popularity and market context

Kimi K3 ranked #13 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

#13

Global market share

Not inferred

Observation date

2026-09-06

OpenAI-compatible API example

This example applies to the published Kendr alias on this hosted profile. Check the live public catalog before use.

curl https://api.kendr.org/v1/chat/completions \
  -H "Authorization: Bearer $KENDR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kc-kimi-k3","messages":[{"role":"user","content":"Hello"}]}'

Frequently asked questions

Sources and evidence dates

1. Kimi K3 third-party catalog record (catalog, checked 2026-09-06)
2. Third-party model catalog methodology (methodology, checked 2026-09-06)
3. Moonshot AI official model documentation (primary, checked 2026-09-06)
4. Artificial Analysis capability indices methodology (benchmark, checked 2026-09-06)
5. Design Arena leaderboard and methodology (benchmark, checked 2026-09-06)
6. Kimi K3 third-party catalog record (benchmark, checked 2026-09-06)
7. Kimi K3 provider or route reference (primary, checked 2026-09-06)