Stay connected with us on X, Discord, and LinkedIn to stay up to date with future analysis
Log in

DeepSeek has launched a newer model, DeepSeek V3.1 Terminus, we suggest considering this model instead.

For more information, see Comparison of DeepSeek V3.1 Terminus to other models and API provider benchmarks for DeepSeek V3.1 Terminus.

DeepSeek V3.1 (Non-reasoning) logo

DeepSeek V3.1 (Non-reasoning) API Provider Benchmarking & Analysis

DeepSeek

Open weights model

Released August 2025

Analysis of API providers for DeepSeek V3.1 (Non-reasoning) across performance metrics including latency (time to first token), output speed (output tokens per second), price and others. API providers benchmarked include Amazon Bedrock, Weights & Biases, Novita, Together.ai, Google Vertex, SambaNova, Fireworks, Baseten (FP8), Lightning AI, DeepInfra (FP4), Eigen AI.

Fastest

#1
FireworksFireworks
289.0 t/s
#2
Eigen AIEigen AI
273.0 t/s
#3
Google VertexGoogle Vertex
265.0 t/s
#4
Together.aiTogether.ai
257.4 t/s
#5
Baseten (FP8)Baseten (FP8)
221.8 t/s

Output speed

Total 11 providers

Lowest Latency

#1
FireworksFireworks
0.55 s
#2
Together.aiTogether.ai
0.59 s
#3
Lightning AILightning AI
0.78 s
#4
Baseten (FP8)Baseten (FP8)
0.89 s
#5
Weights & BiasesWeights & Biases
1.15 s

Time to first token

Total 11 providers

Lowest Price

#1
DeepInfra (FP4)DeepInfra (FP4)
$0.35
#2
Lightning AILightning AI
$0.52
#3
Baseten (FP8)Baseten (FP8)
$0.75
#4
Weights & BiasesWeights & Biases
$0.82
#5
NovitaNovita
$0.83

Blended price (per 1M tokens)

Total 11 providers

DeepSeek V3.1 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 Fireworks (289.0 t/s), Eigen AI (273.0 t/s), Google Vertex (265.0 t/s).
  • For latency, Fireworks (0.55s), Together.ai (0.59s), Lightning AI (0.78s) offer the lowest time to first token.
  • For pricing, DeepInfra (FP4) (0.35), Lightning AI (0.52), Baseten (FP8) (0.75) offer the lowest blended prices per 1M tokens. Prices vary up to 2.3x across providers.
  • Fireworks offers the best performance with both the highest speed and lowest latency. For cost optimization, DeepInfra (FP4) provides the most competitive pricing.
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better

Navigation

PricingSpeedLatencyEnd-to-End Response TimeAPI FeaturesSummary Table of Key Comparison Metrics

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

Pricing

Pricing: Input and Output Prices

USD per 1M Tokens; Lower is better
Input price
Output price

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

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

Speed vs. Price

Output Speed: Output Tokens per Second; Price: USD per 1M Tokens
Most attractive quadrant
Amazon
Baseten (FP8)
DeepInfra (FP4)
Eigen AI
Fireworks
Google Vertex
Lightning AI
Novita
SambaNova
Together.ai
Weights & Biases

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, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).

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)
Amazon
Baseten (FP8)
DeepInfra (FP4)
Eigen AI
Fireworks
Google Vertex
Lightning AI
Novita
SambaNova
Together.ai
Weights & Biases

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, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).

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
Input processing time
'Thinking' time (reasoning models)
Outputting time

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
Amazon logoAmazon
Weights & Biases logoWeights & Biases
Novita logoNovita
Together.ai logoTogether.ai
Google Vertex logoGoogle Vertex
SambaNova logoSambaNova
Fireworks logoFireworks
Baseten (FP8) logoBaseten (FP8)
Lightning AI logoLightning AI
DeepInfra (FP4) logoDeepInfra (FP4)
Eigen AI logoEigen AI

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 DeepSeek V3.1 (Non-reasoning) providers

DeepSeek V3.1 (Non-reasoning) is available through 11 API providers: Amazon, Weights & Biases, Novita, Together.ai, Google Vertex, SambaNova, Fireworks, Baseten (FP8), Lightning AI, DeepInfra (FP4), and Eigen AI. Each provider offers different performance characteristics and pricing.

DeepSeek V3.1 (Non-reasoning) is currently available through 11 API providers that we benchmark and track.

The fastest providers for DeepSeek V3.1 (Non-reasoning) by output speed are Fireworks (289.0 t/s), Eigen AI (273.0 t/s), and Google Vertex (265.0 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 DeepSeek V3.1 (Non-reasoning) are Fireworks (0.55s), Together.ai (0.59s), and Lightning AI (0.78s). Lower latency means faster initial response time.

The most affordable providers for DeepSeek V3.1 (Non-reasoning) by blended price are DeepInfra (FP4) ($0.35 per 1M tokens), Lightning AI ($0.52 per 1M tokens), and Baseten (FP8) ($0.75 per 1M tokens). Blended price uses a 3:1 input to output token ratio.

The providers with the lowest input token pricing for DeepSeek V3.1 (Non-reasoning) are DeepInfra (FP4) ($0.21 per 1M input tokens), Lightning AI ($0.32 per 1M input tokens), and Baseten (FP8) ($0.50 per 1M input tokens).

The providers with the lowest output token pricing for DeepSeek V3.1 (Non-reasoning) are DeepInfra (FP4) ($0.79 per 1M output tokens), Lightning AI ($1.10 per 1M output tokens), and Baseten (FP8) ($1.50 per 1M output tokens).

Prices for DeepSeek V3.1 (Non-reasoning) vary up to 9.5x across providers. The most affordable is DeepInfra (FP4) at $0.35 per 1M tokens, while SambaNova charges $3.38 per 1M tokens.

Output speed for DeepSeek V3.1 (Non-reasoning) varies significantly across providers. Fireworks is the fastest at 289.0 t/s, which is 20.7x faster than DeepInfra (FP4) at 14.0 t/s.

6 of 11 providers support JSON mode for DeepSeek V3.1 (Non-reasoning): Weights & Biases, Google Vertex, SambaNova, Fireworks, Lightning AI, and Eigen AI.

9 of 11 providers support function calling for DeepSeek V3.1 (Non-reasoning): Amazon, Weights & Biases, Novita, Google Vertex, SambaNova, Fireworks, Baseten (FP8), DeepInfra (FP4), and Eigen AI.

For DeepSeek V3.1 (Non-reasoning), Fireworks offers the best performance with highest speed and lowest latency. For cost optimization, DeepInfra (FP4) provides the most competitive pricing.

When choosing a provider for DeepSeek V3.1 (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 DeepSeek V3.1 (Non-reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview