Kimi K2.6 (Non-reasoning) logo
Kimi
•

Open weights model

•

Released April 2026

Kimi K2.6 (Non-reasoning) API Provider Benchmarking & Analysis

Model Comparison

Analysis of API providers for Kimi K2.6 (Non-reasoning) across performance metrics including latency (time to first token), output speed (output tokens per second), price and others. API providers benchmarked include Baseten, Together.ai (FP4), SiliconFlow (FP8), Novita, Fireworks, Parasail (INT4), DeepInfra (FP4), CoreWeave, Microsoft Azure, Nebius, Kimi.

Fastest

#1
CoreWeaveCoreWeave
217.5 t/s
#2
BasetenBaseten
178.9 t/s
#3
FireworksFireworks
107.9 t/s
#4
AzureAzure
100.2 t/s
#5
Together.ai (FP4)Together.ai (FP4)
77.8 t/s

Output speed

Total 11 providers

Lowest Latency

#1
BasetenBaseten
0.52 s
#2
Together.ai (FP4)Together.ai (FP4)
0.68 s
#3
FireworksFireworks
0.72 s
#4
DeepInfra (FP4)DeepInfra (FP4)
0.86 s
#5
CoreWeaveCoreWeave
1.16 s

Time to first answer token

Total 11 providers

Lowest Price

Updated
#1
DeepInfra (FP4)DeepInfra (FP4)
$0.60
#2
NovitaNovita
$0.61
#3
Parasail (INT4)Parasail (INT4)
$0.61
#4
FireworksFireworks
$0.70
#5
CoreWeaveCoreWeave
$0.70

Blended price (per 1M tokens)

Total 11 providers

Kimi K2.6 is available through 11 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.

  • For output speed, the top providers are CoreWeave (217.5 t/s), Baseten (178.9 t/s), Fireworks (107.9 t/s). Speed varies significantly across providers, with a 180% difference between the fastest and slowest.
  • For latency, Baseten (0.52s), Together.ai (FP4) (0.68s), Fireworks (0.72s) offer the lowest time to first token.
  • For pricing, DeepInfra (FP4) (0.60), Novita (0.61), Parasail (INT4) (0.61) offer the lowest blended prices per 1M tokens.

Highlights

Speed

Output tokens per second · Higher is better

End-to-End Response Time

Seconds · Lower is better
Updated

Price

USD per 1M tokens (blended) · Lower is better

Update: Default performance benchmarking workload has updated to 10k input tokens to better reflect production use cases. You can still select different workloads above.

PricingPricing now includes “Cache Hit Price” with new blend ratios.

Pricing: Cache Hit, Input, and Output

Price (USD per M tokens) · Lower is better · 10,000 Input Tokens

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

Price per token included in the request/message sent to the API, represented as USD per million Tokens.

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

  • Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive). Blended price charts use Anthropic cache write price for the input leg.
  • Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
  • OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

Price per token generated by the model (received from the API), represented as USD per million Tokens.

Pricing: Blended Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended) · Lower is better

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

  • Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive). Blended price charts use Anthropic cache write price for the input leg.
  • Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
  • OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

Pricing: Cache Discount

1 - (cache hit price / input price) · Higher is better

Reduction in input token cost due to cache hit relative to input price. Formula: 1 - (Cache Hit Price per Token / Input Token Price), where cache hit price is the first-party cache hit price or the median provider cache hit price. Note that this discount figure does not account for all costs associated with cache hits, such as cache write and storage costs.

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

  • Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive). Blended price charts use Anthropic cache write price for the input leg.
  • Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
  • OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

Speed vs. Price

Blended at 7:2:1 (cache-input-output) · Output speed: output tokens per second · USD per 1M tokens (blended)
Most attractive quadrant
Azure
Baseten
CoreWeave
DeepInfra (FP4)
Fireworks
Kimi
Nebius
Novita
Parasail (INT4)
SiliconFlow (FP8)
Together.ai (FP4)

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

  • Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive). Blended price charts use Anthropic cache write price for the input leg.
  • Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
  • OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

Speed

Measured by Output Speed (tokens per second)

Output Speed

Output tokens per second · Higher is better · 10,000 Input Tokens

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

Latency vs. Output Speed

Latency: seconds to first token received · Output speed: output tokens per second · 10,000 Input Tokens
Most attractive quadrant
Size represents Price (USD per M Tokens)
Azure
Baseten
CoreWeave
DeepInfra (FP4)
Fireworks
Kimi
Nebius
Novita
Parasail (INT4)
SiliconFlow (FP8)
Together.ai (FP4)

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

Latency

Measured by Time (seconds) to First Token

Time to First Token

Seconds to first token received · Lower is better · 10,000 Input Tokens

Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

End-to-End Response Time

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

End-to-End Response Time

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · 10,000 Input Tokens

Seconds to receive a 500 token response. Key components:

  • Input time: Time to receive the first response token
  • Thinking time (only for reasoning models): Time reasoning models spend outputting tokens to reason prior to providing an answer. Amount of tokens based on the average reasoning tokens across a diverse set of 60 prompts (methodology details).
  • Answer time: Time to generate 500 output tokens, based on output speed

For fair comparison, the number of reasoning tokens is standardized across all providers for each model based on the model's representative query token counts.

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

API Features

Function (Tool) Calling & JSON Mode

ModelsFunction callingJSON Mode
Baseten logoBaseten
Together.ai (FP4) logoTogether.ai (FP4)
SiliconFlow (FP8) logoSiliconFlow (FP8)
Novita logoNovita
Fireworks logoFireworks
Parasail (INT4) logoParasail (INT4)
DeepInfra (FP4) logoDeepInfra (FP4)
CoreWeave logoCoreWeave
Azure logoAzure
Nebius logoNebius
Kimi logoKimi

Indicates whether the provider supports function calling in their API. Function calling is also known as 'Tool Calling'.

Indicates whether the provider supports JSON mode in their API. When JSON mode is enabled, the models will always return a valid JSON object.

Context Window

Context window: tokens limit · Higher is better

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

While models have their own context window, in cases this is limited by providers.

Summary Table of Key Comparison Metrics

Frequently Asked Questions

Common questions about Kimi K2.6 (Non-reasoning) providers

Kimi K2.6 (Non-reasoning) is available through 11 API providers: Baseten, Together.ai (FP4), SiliconFlow (FP8), Novita, Fireworks, Parasail (INT4), DeepInfra (FP4), CoreWeave, Azure, Nebius, and Kimi. Each provider offers different performance characteristics and pricing.

Kimi K2.6 (Non-reasoning) is currently available through 11 API providers that we benchmark and track.

The fastest providers for Kimi K2.6 (Non-reasoning) by output speed are CoreWeave (217.5 t/s), Baseten (178.9 t/s), and Fireworks (107.9 t/s). Output speed measures how quickly tokens are generated after the model starts responding.

The providers with the lowest time to first token for Kimi K2.6 (Non-reasoning) are Baseten (0.52s), Together.ai (FP4) (0.68s), and Fireworks (0.72s). Lower latency means faster initial response time.

The most affordable providers for Kimi K2.6 (Non-reasoning) by blended price are DeepInfra (FP4) ($0.60 per 1M tokens), Novita ($0.61 per 1M tokens), and Parasail (INT4) ($0.61 per 1M tokens). Blended price uses a 7:2:1 cache hit/input/output token ratio.

The providers with the lowest input token pricing for Kimi K2.6 (Non-reasoning) are Parasail (INT4) ($0.75 per 1M input tokens), DeepInfra (FP4) ($0.75 per 1M input tokens), and Novita ($0.80 per 1M input tokens).

The providers with the lowest output token pricing for Kimi K2.6 (Non-reasoning) are Novita ($3.40 per 1M output tokens), Parasail (INT4) ($3.50 per 1M output tokens), and DeepInfra (FP4) ($3.50 per 1M output tokens).

Prices for Kimi K2.6 (Non-reasoning) vary up to 2.1x across providers. The most affordable is DeepInfra (FP4) at $0.60 per 1M tokens, while Azure charges $1.25 per 1M tokens.

Output speed for Kimi K2.6 (Non-reasoning) varies significantly across providers. CoreWeave is the fastest at 217.5 t/s, which is 88.0x faster than Nebius at 2.5 t/s.

All 11 providers of Kimi K2.6 (Non-reasoning) support JSON mode for structured output.

All 11 providers of Kimi K2.6 (Non-reasoning) support function calling (tool use).

The best provider for Kimi K2.6 (Non-reasoning) depends on your priorities: CoreWeave offers the highest output speed, Baseten has the lowest latency, and DeepInfra (FP4) provides the most competitive pricing.

When choosing a provider for Kimi K2.6 (Non-reasoning), consider: output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and API features like JSON mode or function calling.

Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.

For information about Kimi K2.6 (Non-reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview