Kimi K3 API pricing
Moonshot · Premium price tier · Kimi K3 License · released 2026-06. Prices in USD per 1 million tokens. Data snapshot: 2026-11-12.
Outlier listing. The price we quote is one host's rate with input near free and output at or above the going rate, not what most hosts charge, so a cost built on it can be far from what you would pay. The typical host charges about $3.00 per 1M input tokens for this model.
A support ticket workload of 100K requests a month costs $828.80 on Kimi K3. The price we quote for Kimi K3 has changed 26 times since 2026-08-15, most recently by +6.6% on 2026-11-04; it is now the 2nd cheapest of 30 models in the Premium price tier. Buying from Wafer rather than at the price we quote lowers the token price by about 40%, for a quantized mxfp4 build. It is rated a stable release but does not meet all four criteria of the FinOps Friendly badge.
- Input: $0.192 per 1M tokens
- Output: $16.00 per 1M tokens
- Batch API (in/out): Not published
- Prompt-cache read: $0.182 per 1M tokens
- Context window: 1M tokens
- Arena ELO: 1475 as of 2026-11-08
Available on
Direct API, AWS Bedrock, OpenRouter
AWS Bedrock regions: ap-northeast-1, ap-southeast-1, eu-central-1, eu-west-1, us-east-1, us-east-2, us-west-2
Curated by hand; check with the provider before committing.
Market position
On blended price ($11.2576 per 1M tokens at 30% input, 70% output) it is cheaper than 97% of models in the Premium price tier: 2nd cheapest of 30. Across every price tier it is cheaper than 9% of priced text models in the catalogue: 275th cheapest of 303. Its Arena ELO of 1475 ranks 20th of the 160 models that carry one, tied with one other, above 87% of them. With an Arena score of 1475, Kimi K3 clears OptimNow's Strong and Standard working quality thresholds: Strong (Arena 1470+), used for support ticket, knowledge Q&A and marketing content; Standard (Arena 1420+), used for meeting summary, invoice processing and call summary. Each threshold is a distance below the best ranked model, which scores 1530 on Arena scores as of 8 Nov 2026.
Licence, size and capabilities
Kimi K3 is open weights, which is not the same as open source: the weights are published, but under Kimi K3 License, a licence that restricts some uses. Read it before self-hosting commercially. Its published size is 2.8T parameters. Listed capabilities: Text, Vision, Reasoning, Agents. Filling its 1M context window once costs $0.192 in input tokens at list price.
Pricing by host
A host is a company that runs the model and sells it by the token: the model maker, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent host. One model, several hosts, and often several prices. This model is sold by 18 different hosts. The price we quote elsewhere on this page is marked. We quote one host's rate so that models stay comparable; it is not a recommendation to buy there.
Cheapest host: Wafer at $0.276 input / $9.60 output per 1M tokens, 40% below the price we quote on the blended price (30% input, 70% output). It serves a quantized mxfp4 build, which is not the same product as the full-precision model. A support ticket workload at 100K requests a month costs $521.40 there against $828.80 at the price we quote, a difference of $307.40 a month. The dearest, Alibaba, charges 1.9x the cheapest.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| Wafer | $0.276 | $9.60 | mxfp4 |
| Chutes | $1.85 | $9.50 | mxfp4 |
| Relace | $0.1824 | $11.0055 | fp4 |
| DigitalOcean | $2.55 | $12.95 | not declared |
| Io Net | $2.30 | $14.00 | fp4 |
| InferenceNet | $0.234 | $15.00 | fp4 |
| DeepInfra | $2.85 | $14.25 | mxfp4 |
| Moonshot AI (model maker) | $3.00 | $15.00 | not declared |
| KT | $3.00 | $15.00 | not declared |
| Amazon Bedrock | $3.00 | $15.00 | not declared |
| Parasail | $3.00 | $15.00 | fp4 |
| Modal | $3.00 | $15.00 | mxfp4 |
| Together | $3.00 | $15.00 | not declared |
| Fireworks | $3.00 | $15.00 | not declared |
| BaseTen | $3.00 | $15.00 | fp8 |
| Sail Research | $0.345 | $16.58 | fp4 |
| Makora | $1.20 | $17.00 | fp4 |
| Alibaba | $3.45 | $17.25 | not declared |
List prices as published on 2026-11-12, via OpenRouter. Negotiated rates, committed-use discounts and provisioned capacity are not shown. These hosts do not all serve the model at the same numeric precision, so the cheapest is not a like-for-like substitute for the dearest.
Monthly cost at 100K requests, by business use case
From the after-discounts figure (prompt caching on the cached input share, batch API for async workloads, only where Moonshot publishes those rates), which assumes you implement both, to the list price. One figure means no discount is published.
| Use case | Tokens (in/out) | Monthly (after discounts – list) |
|---|---|---|
| Support Ticket | 1,500 / 500 | $828 – $829 |
| Knowledge Q&A | 2,000 / 800 | $1.3K |
| Meeting Summary | 10,000 / 1,200 | $2.1K |
| Marketing Content | 2,500 / 1,800 | $2.9K |
| Coding Task | 3,000 / 2,000 | $3.3K |
| Invoice Processing | 1,500 / 600 | $988 – $989 |
| Call Summary | 2,000 / 700 | $1.2K |
| Agent Workflow | 6,000 / 3,000 | $4.9K |
Price history
Kimi K3 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $3.00 input / $15.00 output per 1M tokens. The price we quote has changed 26 times since then, 5 of them a change of quoted host rather than a repricing. The latest change, on 2026-11-04, was +6.6% on the blended price (30% input, 70% output), to $0.192 input / $16.00 output. Net of all changes, the blended price is -1.2% against 2026-08-15.
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-11-04 | $0.192 / $15.00 | $0.192 / $16.00 | +6.6% |
| 2026-11-01 | $0.169 / $20.00 | $0.192 / $15.00 | -24.9% |
| 2026-10-28 | $0.159 / $20.00 | $0.169 / $20.00 | under 0.1% |
| 2026-10-27 | $0.191 / $20.00 | $0.159 / $20.00 | -0.1% |
| 2026-10-26 | $0.239 / $10.80 | $0.191 / $20.00 | +84.2% |
21 earlier changes are not listed.
Alternatives to Kimi K3
None of the 4 comparable models below costs less than Kimi K3 on the blended price.
- Next cheaper model in the Premium price tier: GPT-5.4 (OpenAI) at $11.25 blended, under 1% less.
- Next more expensive: Claude Sonnet 4.5 (Anthropic) at $11.40 blended, 1% more.
The nearest models in the Premium price tier on Arena ELO, each with its blended price and score difference against Kimi K3.
- Claude Opus 4.7 (Anthropic): $5.00 in / $25.00 out per 1M, Arena ELO 1483 (69% more expensive, 8 ELO points higher)
- GPT-5.5 (OpenAI): $5.00 in / $30.00 out per 1M, Arena ELO 1466 (100% more expensive, 9 ELO points lower)
- Claude Fable 5 (Anthropic): $10.00 in / $50.00 out per 1M, Arena ELO 1491 (238% more expensive, 16 ELO points higher)
- Claude Sonnet 4.6 (Anthropic): $3.00 in / $15.00 out per 1M, Arena ELO 1458 (1% more expensive, 17 ELO points lower)
- Cheapest within 15 Arena points: GLM 5.3 Flash (Zhipu), 99% cheaper, 4 ELO points lower, out of 18 models in that band.
- Highest Arena ELO within 20% of this price: Kimi K3 itself has the highest score of the 6 models priced within 20% of it.
Kimi K3 head to head
Other models from Moonshot
Moonshot has 7 other models in the catalogue. Prices per 1M tokens, input then output.
- Kimi K2 0711: $0.57 / $2.30
- Kimi K2 0905: $0.60 / $2.50
- Kimi K2 Thinking: $0.60 / $2.50
- Kimi K2.5: $0.532 / $3.325
- Kimi K2.6: $0.65 / $3.41
- Kimi K2.7 Code: $0.60 / $3.50
- Kimi K3.1 Pro: $0.2375 / $8.4085
All 8 Moonshot models and prices
Common questions about Kimi K3 pricing
- How much does Kimi K3 cost?
- We quote Kimi K3 at $0.192 per 1M input tokens and $16.00 per 1M output tokens. That figure matches none of the rates its 18 hosts list. The model maker, Moonshot AI, lists it at $3.00 per 1M input tokens and $15.00 per 1M output tokens. Prices as listed on 2026-11-12.
- Which host is cheapest for Kimi K3?
- Cheapest host: Wafer at $0.276 input / $9.60 output per 1M tokens, 40% below the price we quote on the blended price (30% input, 70% output). It serves a quantized mxfp4 build, which is not the same product as the full-precision model. A support ticket workload at 100K requests a month costs $521.40 there against $828.80 at the price we quote, a difference of $307.40 a month. The dearest, Alibaba, charges 1.9x the cheapest. Host rates as listed on 2026-11-12.
- Does Kimi K3 offer batch or cached pricing?
- Yes, for prompt caching only. Cache reads cost $0.182 per 1M tokens, as listed on 2026-11-12. No batch pricing is published.
- What would Kimi K3 cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $829 on Kimi K3 at the list price quoted on 2026-11-12.
- Can Kimi K3 be self-hosted?
- Yes, within the terms of its licence. Kimi K3 is open weights, which is not the same as open source: the weights are published, but under Kimi K3 License, a licence that restricts some uses. Read it before self-hosting commercially.
- Has the price of Kimi K3 changed?
- Kimi K3 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $3.00 input / $15.00 output per 1M tokens. The price we quote has changed 26 times since then, 5 of them a change of quoted host rather than a repricing. The latest change, on 2026-11-04, was +6.6% on the blended price (30% input, 70% output), to $0.192 input / $16.00 output. Net of all changes, the blended price is -1.2% against 2026-08-15.
Its weights are published, which changes what you may run yourself: the cost gap between open weights and proprietary APIs.
These are list prices. Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Build the business case · Get a FinOps review
LLM pricing · Models by maker · Cloud compute pricing · Compare models · Price barometer · FinOps guides · API and MCP server