Kimi K3.1 Pro is out, with open weights by November 4. How to use it now

What Moonshot released

On October 23, 2026, Moonshot AI announced Kimi K3.1 Pro, “the first model in our new K3.1 series and our most capable 3T-class open-weight model to date,” with a context window of one million tokens. It’s a post-trained version of Kimi K3, Moonshot’s previous large model, made with what the company calls “substantially larger post-training compute.” A second version, Kimi K3.1 Pro Highspeed, runs at “2-3× the speed” of the first.

Both are available now on Kimi.ai and the Kimi mobile apps, in the Kimi Work desktop app, in Kimi Code, Moonshot’s coding agent, and through the Kimi API as kimi-k3.1-pro and kimi-k3.1-pro-highspeed. Artificial Analysis gives its size as 2.8 trillion parameters in total, 104 billion of them active for each token, with text and image input.

The weights aren’t out yet. In Moonshot’s words: “Full model weights will be released by Nov 4, 2026.”

How good it is

Moonshot ran most of its coding tests in Claude Code’s harness, with reasoning effort at its maximum. Here are its own figures against Claude Opus 5.5 and GPT-6 Astra:

Benchmark	Kimi K3.1 Pro	Claude Opus 5.5	GPT-6 Astra
Terminal-Bench 4.0	65.7%	66.4%	58.2%
SWE Marathon	51.7%	46.3%	42.5%
Kimi Code Bench 3.0 Long-horizon	61.2%	58.5%	66.5%
Terminal-Bench Science 0.1	38.1%	63.3%	68.1%

Kimi Code Bench is Moonshot’s internal test, and on it GPT-6 Astra still leads.

Artificial Analysis tested the model independently and ranks it fourth on its Intelligence Index at 55, behind Claude Opus 5.5 at 58, and Claude Sonnet 5.5 and Muse Wonder Full at 56. It scored 67% on Terminal-Bench 4.0, which Artificial Analysis calls “the highest score of any model” on that test. It also uses 55,000 output tokens per index task, 14% more than K3, so a task costs about the same as on K3, $1.98, despite the lower prices. Once the weights are public, Artificial Analysis expects it to be the highest-scoring open-weights model, cutting the gap to the best proprietary model from 12 points to 3. Its Coding Agent Index doesn’t include K3.1 Pro yet.

What it costs

On the API, per million tokens:

Model	Input	Output
Kimi K3.1 Pro	$2	$10
Kimi K3.1 Pro Highspeed	$4	$20
Kimi K3	$3	$15
Claude Sonnet 5.5	$2	$10
Claude Opus 5.5	$4	$20
GPT-6.1 Sol	$2	$10

Kimi’s rates are from its pricing page, where cached input costs $0.20 on both K3.1 models. The others are from Anthropic and OpenAI. K3.1 Pro is a third cheaper than K3, but on price per token it matches Claude Sonnet 5.5 and GPT-6.1 Sol rather than undercutting them.

To use it in Kimi Code without paying per token, you need a Kimi membership. Kimi’s pricing page lists Kimi Code on Plus, Pro, Max and Ultra, at $180, $372, $948 and $1,908 a year, billed annually, and not on the free plan. Pro and above add Highspeed. For new members, Kimi’s membership guide says “the weekly quota limit is removed,” leaving a rolling five-hour window and a monthly total, but it doesn’t publish how much either allows.

OpenCode Go, OpenCode’s subscription to open models, includes K3.1 Pro too. On the $10-a-month Go plan, the model has a monthly limit of $15 of use at API prices, which OpenCode estimates at 147 requests every five hours, 368 a week or 735 a month. Go Plus, at $40, raises the limits.

Can you run it yourself?

Not yet, and not on a laptop. Until the weights arrive, K3.1 Pro is a hosted model only.

Moonshot hasn’t named a license for the weights. Kimi K3’s weights came under Moonshot’s own Kimi K3 License, which allows commercial use with two conditions: a business selling model access with more than $20 million in revenue over 12 months needs a separate agreement with Moonshot, and a product with more than 100 million monthly users or $20 million in monthly revenue must show “Kimi K3” in its interface. Internal use is exempt from both.

On size, Kimi K3 on Hugging Face has 2.8 trillion parameters, and its files add up to about 1.56 TB, far more than any laptop’s memory. Its card names vLLM, SGLang and TokenSpeed to serve it. For a coding model that does fit on a laptop, see Flint.

Moonshot also cut some of what the open model can do. It removed cyber-related data from training for the open-weight release: on xdev-bench-hard, its internal vulnerability-research test, K3.1 Pro solves 81.0% of problems, against 90.5% for the same training without the cuts. It says K3.1 Pro’s cyber capability “remains behind that of frontier proprietary models that have been publicly available for at least three months.”

Four ways to use it in a coding agent today

1. Kimi Code. Moonshot’s own agent comes as a CLI, a desktop app and a VS Code extension. Its docs install the CLI with curl -fsSL https://code.kimi.com/kimi-code/install.sh | bash; run kimi, sign in with /login, and choose K3.1 Pro with /model. It needs the Plus membership or above.
2. Claude Code. Kimi’s guide points Claude Code at Kimi with a Kimi Code membership key: set ANTHROPIC_BASE_URL to https://api.kimi.ai/coding/, ANTHROPIC_API_KEY to the key, and ANTHROPIC_MODEL and Claude Code’s default model variables to k3.1-pro. /status then shows Kimi’s address. Every request goes to Kimi instead of Anthropic, and thinking has to stay on: K3.1 Pro rejects requests without it.
3. Codex. Kimi’s Codex guide adds a provider to ~/.codex/config.toml with base_url = "https://api.moonshot.ai/v1" and wire_api = "responses", and reads a Kimi API key from KIMI_API_KEY. You pay per token.
4. OpenCode. Run /connect, choose Moonshot AI, enter a key from Kimi’s API platform, and pick the model with /models, as OpenCode’s provider guide describes. Models.dev, the catalog OpenCode takes its model lists from, already lists Kimi K3.1 Pro under Moonshot AI and OpenCode Go.

Enjoy runs OpenCode beside Claude Code, Codex and the other labs’ agents. Add Moonshot AI to OpenCode once, with opencode auth login in a terminal, and Kimi K3.1 Pro appears in Enjoy’s model picker with the rest of OpenCode’s models. Hand it a conversation when a Claude or ChatGPT plan runs low, or give it the same task to compare. Enjoy itself is free; Kimi bills you for the tokens, and Enjoy doesn’t add Kimi to a Claude or ChatGPT account.

What Moonshot hasn’t said

• The license for K3.1 Pro’s weights, or whether Highspeed’s weights come too. The announcement promises “full model weights” without naming the model they belong to.
• What it takes to run, in memory or GPUs.
• How much each Kimi plan includes in Kimi Code. The membership guide describes the windows, not their size.
• Independent coding results. Most of the coding figures above are Moonshot’s own; Artificial Analysis is so far the main outside check.

We’ll update this memo when the weights are published.