moonshotai
MoonshotAI: Kimi K3.1 Pro (batch)
moonshotai/kimi-k3.1-pro:batch
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
MoonshotAI's Kimi K3.1 Pro (batch) is an open-weight, natively multimodal reasoning model exposed through OpenRouter's batch variant, accepting text, image and, per the catalog, video input and returning text. Post-trained from Kimi K3, it uses a large mixture-of-experts architecture and supports a context window of about one million tokens. The vendor describes it as aimed at long-horizon agentic and coding work; configurable reasoning effort is supported and enabled by default. Vendor-reported and arena evaluations are positive in several categories, but independent, workload-specific validation is still needed.[1][2]
Strengths
- Open-weight mixture-of-experts design with sparse activation, described by the vendor as its most capable model in this class to date.[2]
- Very long context window of roughly one million tokens, suited to lengthy documents and extended agent sessions.[1][2]
- Multimodal input and configurable reasoning effort, with vendor-published and arena evaluations reporting competitive coding and agentic results.[1][2]
Limitations
- Evidence comes only from a marketplace catalog and the vendor's own model card; reported scores are vendor-run or arena-based and may not predict results on your workloads, so validate on real tasks before adopting.[1][2]
- Sources disagree on input scope: the model card lists text and image, while OpenRouter's catalog also lists video; video support is unverified here.[1][2]
- Knowledge cutoff, batch turnaround and latency behavior, and the custom kimi-k3.1-pro license terms are not documented in this source pack.[1][2]
Where to use it
- Batch processing of long documents and large codebases that may benefit from the extended context window.[1][2]
- Agentic coding and app, web, or design-oriented generation, which vendor and arena evaluations emphasize.[1][2]
- Reasoning-heavy research or analysis pipelines where configurable effort and batch turnaround are acceptable trade-offs.[1][2]
Sources
- [1] OpenRouter catalog · MoonshotAI: Kimi K3.1 Pro (batch)Sources checked: Nov 13, 2026
- [2] Model documentation · MoonshotAI: Kimi K3.1 Pro (batch)Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...
Specifications
- Model ID
- moonshotai/kimi-k3.1-pro:batch
- Canonical ID
- moonshotai/kimi-k3.1-pro-20261023
- Creator
- moonshotai
- Added to OpenRouter
- 23.10.2026
- 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 $0.700 / out $6
- 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 | AA — next 20% of the cohort | reasoning: optional · default max · efforts: max, high, low, none · parameters: include_reasoning, reasoning, reasoning_effort |
| Coding | No measurement | — |
| Agents | No measurement | — |
| 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 | 46.5 | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · androidnativewin rate 42.6% · rank 28 | 1,149Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · codetovideowin rate 58.5% · rank 5 | 1,232Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · fullstackwin rate 61.9% · rank 2 | 1,327Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · mobileappswin rate 59.7% · rank 4 | 1,273Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · pdfwin rate 57.3% · rank 12 | 1,226Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · python-pptxslideswin rate 67.3% · rank 3 | 1,292Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · webappswin rate 61% · rank 3 | 1,307Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · 3dwin rate 70.8% · rank 3 | 1,447Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · asciiartwin rate 59.1% · rank 6 | 1,302Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · codecategorieswin rate 60.8% · rank 7 | 1,351Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · datavizwin rate 56% · rank 13 | 1,304Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · gamedevwin rate 66.1% · rank 4 | 1,406Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · svgwin rate 62.6% · rank 2 | 1,329Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · uicomponentwin rate 65.4% · rank 1 | 1,388Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · websitewin rate 57.9% · rank 6 | 1,323Elo | 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) |
|---|---|---|---|---|---|---|
| Sail Researchfp4 | 1,048,576 | $0.700 | $6 | — | — | — |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.