BroadRouter

Kimi K3.1 Pro vs Kimi K3 API pricing

Updated

Kimi K3.1 Pro: USD 1.50 input, USD 10 output per 1M tokens (Sail Research). Kimi K3: USD 1.85 input, USD 9.50 output per 1M tokens (Chutes). At 3 input tokens per output token, Kimi K3 costs 3.8% more than Kimi K3.1 Pro (USD 3.763 vs USD 3.625 per 1M).

Kimi K3.1 Pro and Kimi K3 are releases in the same family. Price only: every number is computed from prices read from each provider's own pricing page (linked per cell, newest check 2026-11-16), with no claim about quality, speed or accuracy.

How do Kimi K3.1 Pro and Kimi K3 differ on price?

Kimi K3.1 Pro and Kimi K3 prices side by side

USD per 1M tokens, standard tier. "Cheapest" is the lowest price at 3 input tokens to 1 output token; each line can come from a different provider.

Kimi K3.1 Pro and Kimi K3 API prices compared
PriceKimi K3.1 ProKimi K3
Model pageKimi K3.1 Pro
Moonshot AI · 20 providers
Kimi K3
Moonshot AI · 23 providers
Cheapest standard price, blended 3:1USD 3.625 lower
USD 1.50 in / USD 10 out at Sail Research
Sail Research page checked 2026-11-16
USD 3.763
USD 1.85 in / USD 9.50 out at Chutes
Chutes page checked 2026-11-15
Cheapest inputUSD 1.50 lower
Sail Research source
USD 1.85
Chutes source
Cheapest cached inputUSD 0.18 lower promo
AkashML source
USD 0.185
Chutes source
Cheapest outputUSD 9.50 tie promo
DeepInfra source
USD 9.50 tie
Chutes source
Maker's own APIUSD 2 / USD 10
Moonshot AI source
USD 3 / USD 15
Moonshot AI source
Median standard priceUSD 2 / USD 10
input / output, 20 prices
USD 3 / USD 15
input / output, 23 prices
Context window1,048,576 tokens, largest stated; from 262,144 at some providers1,048,576 tokens, largest stated; from 262,144 at some providers
Providers with a standard price2023

Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).

What do Kimi K3.1 Pro and Kimi K3 cost for a real workload?

30 days at each model's cheapest standard price for the workload (presets of the cost calculator; cached input at the cached price where listed).

Monthly cost of three workloads on Kimi K3.1 Pro and Kimi K3
WorkloadKimi K3.1 ProKimi K3Difference
Support chatbot
2,000 requests a day, 1,500 input tokens (60% cached), 300 output
USD 244.80
Sail Research
USD 247.59
Chutes
USD 2.79
Kimi K3.1 Pro lower
RAG over documents
1,000 requests a day, 8,000 input tokens (30% cached), 500 output
USD 416.40
Sail Research
USD 466.62
Chutes
USD 50.22
Kimi K3.1 Pro lower
Coding agent (one developer)
1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning
USD 1,611
Sail Research
USD 1,638
Chutes
USD 26.55
Kimi K3.1 Pro lower

Kimi K3.1 Pro is cheaper on all three workloads. The widest gap is on RAG over documents: the dearer model costs 12% more than the cheaper one.

Does the cheaper model change with your input:output ratio?

The cheaper model changes with the mix: below 1.4 input tokens per output token, Kimi K3 is cheaper; above 1.4, Kimi K3.1 Pro is cheaper. At the site's 3:1 blend, Kimi K3.1 Pro is cheaper. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.

Price per 1M tokens of Kimi K3.1 Pro and Kimi K3 at different input:output mixes
Input : output tokensKimi K3.1 ProKimi K3
1 : 3 (long answers)USD 7.60
DeepInfra
USD 7.588 lower
Chutes
1 : 1USD 5.70
DeepInfra
USD 5.675 lower
Chutes
3 : 1 (the site’s blend)USD 3.625 lower
Sail Research
USD 3.763
Chutes
10 : 1 (retrieval, long prompts)USD 2.273 lower
Sail Research
USD 2.545
Chutes
30 : 1 (agents, big contexts)USD 1.774 lower
Sail Research
USD 2.097
Chutes

USD per 1M tokens at the cheapest standard provider for each mix, before cache discounts. Thinking tokens bill as output.

What do Kimi K3.1 Pro and Kimi K3 cost per character in other languages?

Cost to send 1M characters at each model's cheapest non-promotional standard input price, with the Kimi K3 tokenizer for Kimi K3.1 Pro (the same one for both, so the gap equals the price gap):

Input cost per 1M characters by language, Kimi K3.1 Pro and Kimi K3
ContentKimi K3.1 ProKimi K3
EnglishUSD 0.326 lower
Sail Research
USD 0.402
Chutes
Arabic MSAUSD 0.719 lower
Sail Research
USD 0.887
Chutes
Egyptian ArabicUSD 0.723 lower
Sail Research
USD 0.892
Chutes
HindiUSD 0.852 lower
Sail Research
USD 1.05
Chutes
ChineseUSD 0.818 lower
Sail Research
USD 1.01
Chutes
SpanishUSD 0.440 lower
Sail Research
USD 0.542
Chutes
CodeUSD 0.389 lower
Sail Research
USD 0.480
Chutes

From the real cost by language dataset (measured tokens per character on FLORES-200 and other texts). Your own text will differ.

Check it on your own traffic

More comparisons: Kimi K3 vs Kimi K2.6 · Claude Sonnet 5.5 vs Kimi K3 · Qwen3.8 Max vs Kimi K3 · GLM 5.3 vs Kimi K3 · DeepSeek V4 Pro vs Kimi K3 · Kimi K3 vs MiniMax M3 · every Kimi K3.1 Pro comparison · every Kimi K3 comparison · all 662 price comparisons

Questions

Is Kimi K3.1 Pro cheaper than Kimi K3?

Yes. On a blend of 3 input tokens to 1 output token, Kimi K3.1 Pro costs USD 3.625 per 1M tokens and Kimi K3 USD 3.763, at each model's cheapest standard price. For input alone, Kimi K3.1 Pro is cheaper (USD 1.50 against USD 1.85 per 1M); for output alone, both start at USD 9.50. This compares price only, not quality.

What do Kimi K3.1 Pro and Kimi K3 cost per 1M tokens?

Kimi K3.1 Pro: from USD 1.50 input and USD 10 output at Sail Research; Moonshot AI's own API charges USD 2 / USD 10; median of 20 standard prices USD 2 / USD 10. Kimi K3: from USD 1.85 input and USD 9.50 output at Chutes; Moonshot AI's own API charges USD 3 / USD 15; median of 23 standard prices USD 3 / USD 15. Each price is read from the provider's own pricing page; the source is linked in the table.

Which providers sell both Kimi K3.1 Pro and Kimi K3?

17 providers have a standard price for both: Featherless (USD 4 and USD 4 at 3:1), Relace (USD 4 and USD 5 at 3:1), DeepInfra (USD 3.80 and USD 5.70 at 3:1), Alibaba Cloud (USD 4 and USD 6 at 3:1), Baseten (USD 4 and USD 6 at 3:1), DigitalOcean (USD 4 and USD 6 at 3:1), Doubleword (USD 4 and USD 6 at 3:1), Fireworks AI (USD 4 and USD 6 at 3:1), GMI Cloud (USD 4 and USD 6 at 3:1), Inference.net (USD 4 and USD 6 at 3:1), Modal (USD 4 and USD 6 at 3:1), Moonshot AI (USD 4 and USD 6 at 3:1), Novita AI (USD 4 and USD 6 at 3:1), Parasail (USD 4 and USD 6 at 3:1), River (USD 4 and USD 6 at 3:1), Together AI (USD 4 and USD 6 at 3:1), Venice (USD 4 and USD 7.50 at 3:1). The first is the lowest combined price. Buying each model from its own cheapest provider costs USD 3.625 and USD 3.763.

Which is cheaper for a coding agent or a chatbot, Kimi K3.1 Pro or Kimi K3?

For a coding agent (1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning): USD 1,611 a month on Kimi K3.1 Pro (Sail Research) against USD 1,638 on Kimi K3 (Chutes), a difference of USD 26.55. For a support chatbot (2,000 requests a day, 1,500 input tokens (60% cached), 300 output): USD 244.80 a month on Kimi K3.1 Pro (Sail Research) against USD 247.59 on Kimi K3 (Chutes), a difference of USD 2.79. Each uses the cheapest standard price for that workload, cached input included. Try your own numbers.

At what input:output ratio does the cheaper of Kimi K3.1 Pro and Kimi K3 change?

The cheaper model changes with the mix: below 1.4 input tokens per output token, Kimi K3 is cheaper; above 1.4, Kimi K3.1 Pro is cheaper. At the site's 3:1 blend, Kimi K3.1 Pro is cheaper. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.

Do Kimi K3.1 Pro and Kimi K3 have cached input pricing?

Kimi K3.1 Pro from USD 0.18 per 1M cached input tokens (AkashML; 20 of 20 standard prices list a discount). Kimi K3 from USD 0.185 per 1M cached input tokens (Chutes; 22 of 23 standard prices list a discount). Cached tokens are billed at the input price where no cache price is published.

Which has the larger context window, Kimi K3.1 Pro or Kimi K3?

Both go up to 1,048,576 tokens on the providers that state it.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity