Kimi K3.1 Pro is an open-weight multimodal chat model from Moonshot AI that processes text and image inputs to generate text outputs. It is designed for tasks such as coding, financial research, and agentic workflows, supporting capabilities like calling web search tools and structured output generation.
Input
Output
Context
1M
Max Output
-
Parameters
2780.1B
Input Modalities
Output Modalities
Bands compare this model with every model the same suite measured in our catalog.
Download 1561 GB · 8 GGUF builds
| Precision | Weights | One request at 33K tokens | Smallest GPU that fits |
|---|---|---|---|
| FP16 / BF16 | 5178 GB | 5476 GB | 16x AMD Instinct MI455X (6912 GB) |
| INT8 | 2589 GB | 2756 GB | 8x AMD Instinct MI455X (3456 GB) |
| INT4 | 1295 GB | 1396 GB | 4x AMD Instinct MI455X (1728 GB) |
Computed by Inferbase from the 2,780,110,363,392 parameters in the creator's Hugging Face repository; one request adds its KV cache, activations and framework overhead, with the layer layout estimated from the parameter count.
Public resources
Specification sources
As of Nov 21, 2026.
Answered from the facts on this page, with their sources.
Kimi K3.1 Pro has a context window of 1,048,576 tokens (source: Creator or provider API). A provider may serve less; Inferbase routes on what each provider serves.
Yes. Kimi K3.1 Pro is an open-weight model under the Other license. You can download and self-host it within the license terms.
Kimi K3.1 Pro has 2,780,110,363,392 parameters (2780.1B), as stated in the creator's Hugging Face repository.
Strengths: 1,048K token context window. Each is computed from the model's stated facts and benchmark results.
Kimi K3.1 Pro accepts Text and Image and produces Text. The Capabilities card lists the API features it supports, such as function calling and structured output.
Kimi K3.1 Pro has published results from Epoch AI. The Benchmarks card shows each score as reported, with its band among the models the same source measured. Scores are never combined across sources.
Yes. Kimi K3.1 Pro is available on Inferbase Managed through the OpenAI-compatible API as moonshot-kimi-k3-1-pro. Request it by name, or request auto and routing picks it when it fits the request best.
The weights take 5.18 TB at FP16 / BF16, 2.59 TB at INT8, 1.29 TB at INT4, computed from the 2,780,110,363,392 parameters its creator publishes. Each request adds its KV cache and the framework's overhead. The hardware section shows one request and the smallest GPU that fits; the GPU planner sizes it for your context length and traffic.
Start on the free plan with 5,000 requests a month, no card required.