moonshotai
MoonshotAI: Kimi K3 (batch)
moonshotai/kimi-k3:batch
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
moonshotai/kimi-k3:batch is the OpenRouter batch variant of Moonshot AI's Kimi K3. The catalog and model documentation describe an open-weight multimodal MoE model with 2.8T total and 104B activated parameters, text generation, image and video inputs, and a 1,048,576-token context window. Moonshot positions it for long-horizon coding, knowledge work, and agentic workflows. The catalog reports adjustable reasoning effort, tool use, and structured-output parameters; batch-specific behavior and limits are not documented here.[1][2]
Strengths
- The catalog lists a 1,048,576-token context window and text, image, and video inputs.[1]
- Moonshot describes an MoE design activating 104B of 2.8T parameters for extended engineering and agentic sessions.[2]
- OpenRouter lists reasoning-effort control, tool calling, response formats, and structured outputs.[1]
Limitations
- The supplied evidence does not define latency, throughput, retention, or service limits specific to the :batch variant.[1][2]
- The model-card summary lists text and image, while its prose and OpenRouter describe video input; endpoint validation is warranted.[1][2]
- Vendor and catalog results vary by category and may not transfer; validate against the intended workload.[1][2]
Where to use it
- Consider it for agentic coding across large repositories when tools and long context are needed.[1][2]
- It may fit multimodal analysis combining text with image or video inputs, subject to endpoint validation.[1][2]
- It could support batch-oriented knowledge work, although the supplied evidence does not define this variant's batch semantics.[1][2]
Sources
- [1] OpenRouter catalog · MoonshotAI: Kimi K3 (batch)Sources checked: Nov 13, 2026
- [2] Model documentation · MoonshotAI: Kimi K3 (batch)Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
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...
Specifications
- Model ID
- moonshotai/kimi-k3:batch
- Canonical ID
- moonshotai/kimi-k3-20260715
- Creator
- moonshotai
- Added to OpenRouter
- 16.07.2026
- Context length
- 1,048,576
- Modality
- text+image+video->text
- Input modalities
- text, image, video
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $2.28 / out $11.4
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | CC — below the B band | reasoning: optional · default max · efforts: max, high, low, none · parameters: include_reasoning, reasoning, reasoning_effort |
| Coding | AA — next 20% of the cohort | — |
| Agents | AA — next 20% of the cohort | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · default max · efforts: max, high, low, none · parameters: include_reasoning, reasoning, reasoning_effort |
| Document analysis | No measurement | — |
| Vision | No measurement | input: image |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | input: video |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Intelligence index | 17.4 | CC — below the B band | 13.11.2026 |
| Artificial Analysis | Coding index | 72 | AA — next 20% of the cohort | 13.11.2026 |
| Artificial Analysis | Agentic index | 29.3 | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · agenticgamedevwin rate 54% · rank 4 | 1,202Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · androidnativewin rate 57.7% · rank 7 | 1,255Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · blender3dwin rate 40.1% · rank 9 | 1,156Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · codetovideowin rate 63.3% · rank 3 | 1,269Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · fullstackwin rate 61.5% · rank 4 | 1,295Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · godotgamedevwin rate 48.5% · rank 14 | 1,204Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · htmlslideswin rate 55.6% · rank 5 | 1,152Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · mobileappswin rate 56.7% · rank 8 | 1,256Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · pdfwin rate 66% · rank 6 | 1,278Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · python-pptxslideswin rate 54.7% · rank 9 | 1,220Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · spreadsheetwin rate 33.5% · rank 19 | 1,128Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · webappswin rate 59.3% · rank 7 | 1,280Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · 3dwin rate 63.8% · rank 10 | 1,367Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · codecategorieswin rate 60.3% · rank 6 | 1,352Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · datavizwin rate 63.6% · rank 4 | 1,351Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · gamedevwin rate 54.7% · rank 9 | 1,353Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · svgwin rate 59.7% · rank 7 | 1,298Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · uicomponentwin rate 60.5% · rank 10 | 1,339Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · websitewin rate 59.3% · rank 4 | 1,337Elo | SS — top 10% of the cohort | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| DeepInfrabf16 | 1,048,576 | $2.28 | $11.4 | — | — | — |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.