Kimi
This model always redirects to the latest model in the Kimi family.
kimi-latestReleased 27 Apr 2026 (Apr 2026)Pricing
List prices at Fireworks AI, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $2.00 | $10.00 | |
| Cached input 90% off input | $0.20 | 90% off input | |
| Priority 50% more | $3.00 | $15.00 | 50% more |
What it costs a month
At Fireworks AI list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$1,188
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$888
- Coding agent1,000 requests a day, 40K in / 2K out, cached$1,488
- Overnight batch20,000 requests a day, 4K in / 0.4K out$7,200
Where to run it
1 host serve Kimi. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
Fireworks AIInference kimi-latest | $2.00 / $10.00 | kimi-latest |
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- TextImages
- Produces
- Text
- Effort settings the source lists
- lowhighmax
- Features
- Prompt caching
- Open weights
- License: Modified MIT LicenseRuntimes: SGLang, vLLM
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than Fireworks AI.
Smaller Moonshot AI models
No cheaper Moonshot AI model keeps its key capabilities.
Other frontier labs
- 25% less
- 25% less
- 50% less
- 95% less
Open-weight picks
- 0% less
- 47% less
- 57% less
- 87% less
Call it
Using Fireworks AI's API.
from litellm import completion
response = completion(
model="fireworks_ai/kimi-latest",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does Kimi cost?
On Fireworks AI, Kimi costs $2.00 per million input tokens and $10.00 per million output tokens. Cached input costs $0.20 per million tokens.
Which APIs offer Kimi?
Kimi runs on 1 API: Fireworks AI.
What is Kimi's context window?
1M tokens, with up to 131K tokens in one reply.
Is Kimi being retired?
No retirement date is scheduled for it on any API we track.
What are cheaper alternatives to Kimi?
Models that cost less and keep most of what it can do: Gemini 4 Hydrogen on Google AI Studio (25% less), Grok 4.7 on xAI (25% less) and Kimi K3.1 Pro on Databricks (0% less). Savings are on a mix of 3 input tokens to 1 output token.
When was Kimi released?
Moonshot AI released it on 27 Apr 2026.
What is Kimi's API model id?
Use kimi-latest with Fireworks AI.
Sources
Raw catalog entry for kimi-latest on Fireworks AI
{
"cache_read_input_token_cost": 2e-7,
"cache_read_input_token_cost_priority": 3e-7,
"input_cost_per_token": 0.000002,
"input_cost_per_token_priority": 0.000003,
"litellm_provider": "fireworks_ai",
"max_input_tokens": 1048576,
"max_output_tokens": 131072,
"mode": "chat",
"output_cost_per_token": 0.00001,
"output_cost_per_token_priority": 0.000015,
"reasoning_effort_levels": [
"low",
"high",
"max"
],
"source": "https://api.fireworks.ai/v1/serverless/models?format=nested",
"supports_function_calling": true,
"supports_mid_conversation_system": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_vision": true
}