DeepSeek Chat

DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations reveal that the model outperforms other open-source models and rivals leading closed-source models.

For model details, please visit the DeepSeek-V3 repo for more information, or see the launch announcement.

deepseek/deepseek-chat

Organization: DeepSeek
Family: deepseek
Providers: 6
Context: 1,000,000
Output limit: 384,000
Knowledge: 2025-09
Release: 2025-12-01
Updated: 2026-02-28
Weights: Open
Input: text
Output: text
Capabilities: Tools, Temperature

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

7 offers · USD per million tokens

Provider	Model ID	Context	Output	Input / output · 1M	Reasoning	Tools	Structured	Details
AnyAPI	deepseek/deepseek-chat	1,000,000	384,000	— / —	No	Yes	—	View
Eden AI	deepseek/deepseek-chat	131,072	384,000	$0.15 / $0.60	No	Yes	Yes	View
Kilo Gateway	deepseek/deepseek-chat	128,000	16,000	$0.2574 / $1.0287	No	Yes	Yes	View
NanoGPT	deepseek-chat	128,000	8,192	$0.10 / $0.425	No	Yes	Yes	View
DeepInfravia OpenRouter	deepseek/deepseek-chat fp4	163,840	16,384	$0.32 / $0.89	No	Yes	Yes	View
StreamLakevia OpenRouter	deepseek/deepseek-chat unknown	128,000	16,000	$0.2574 / $1.0287	No	Yes	Yes	View
OrcaRouter	deepseek/deepseek-chat	1,000,000	384,000	$0.147 / $0.295	No	Yes	—	View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens

Input from$0.10Per 1M input tokens

Output from$0.295Per 1M output tokens

Available offers7Across 6 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 2 lowest-priced configurations

InputOutput

StreamLakestreamlake

$0.2574$1.0287

DeepInfradeepinfra/fp4

$0.32$0.89

deepseek/deepseek-chatObserved Oct 4, 2026, 11:31 AM UTC

Input: $0.2574
Output: $1.0287

Hosting prices through OpenRouter All endpoints and cache rates

Endpoint	Input / 1M	Output / 1M	Cache read / 1M	Cache write / 1M
DeepInfradeepinfra/fp4	$0.32	$0.89	—	—
StreamLakestreamlake	$0.2574	$1.0287	—	—

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 2 endpoints

P50P75P90P99

DeepInfradeepinfra/fp4

484–4,804.83 ms

StreamLakestreamlake

1,407–5,933.62 ms

All latency percentiles 2 endpoints

Endpoint	P50	P75	P90	P99
DeepInfradeepinfra/fp4	484 ms	723 ms	1,265.8 ms	4,804.83 ms
StreamLakestreamlake	1,407 ms	1,757.25 ms	2,485.8 ms	5,933.62 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 2 endpoints

P50P75P90P99

DeepInfradeepinfra/fp4

28–65 t/s

StreamLakestreamlake

40–70 t/s

All throughput percentiles 2 endpoints

Endpoint	P50	P75	P90	P99
DeepInfradeepinfra/fp4	28 t/s	34 t/s	41 t/s	65 t/s
StreamLakestreamlake	40 t/s	48 t/s	56 t/s	70 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window

Availability by endpointCurrent overlapping windows · 2 endpoints

Endpoint5 minutes30 minutes24 hours

DeepInfradeepinfra/fp4Observed Oct 4, 2026, 11:31 AM UTC

100%

99.85%

95.53%

StreamLakestreamlakeObserved Oct 4, 2026, 11:31 AM UTC

99.1%

99.41%

99.24%

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations

deepseek/deepseek-chatObserved Oct 4, 2026, 11:31 AM UTC

Design Arena	Category	Elo	Win rate	Rank
models	3d	1,102	50.5%	82
models	codecategories	1,114	48.4%	91
models	dataviz	1,097	50%	95
models	gamedev	1,066	43.7%	98
models	svg	975	38.8%	87
models	uicomponent	1,100	52.7%	89
models	website	1,126	48.5%	91

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better

gpqa diamondDeepSeek: DeepSeek V3

62.121%

tau bench verified airlineDeepSeek: DeepSeek V3

39.444%

Evaluation	Configuration	Source	Results	Details
Composite indices	DeepSeek V3 (Dec '24)deepseek/deepseek-chat	Artificial Analysis	Intelligence 8.5 · Coding 23 · Agentic 0.8	Details
models · uicomponent	DeepSeek-V3-0324deepseek/deepseek-chat-v3	Design Arena	1,100 Elo	Details
models · 3d	DeepSeek-V3-0324deepseek/deepseek-chat-v3	Design Arena	1,102 Elo	Details
models · website	DeepSeek-V3-0324deepseek/deepseek-chat-v3	Design Arena	1,126 Elo	Details
models · dataviz	DeepSeek-V3-0324deepseek/deepseek-chat-v3	Design Arena	1,097 Elo	Details
gpqa diamond	DeepSeek: DeepSeek V3deepseek/deepseek-chat-v3	OpenRouter	62.121%	Details
models · svg	DeepSeek-V3-0324deepseek/deepseek-chat-v3	Design Arena	975 Elo	Details
models · codecategories	DeepSeek-V3-0324deepseek/deepseek-chat-v3	Design Arena	1,114 Elo	Details
tau bench verified airline	DeepSeek: DeepSeek V3deepseek/deepseek-chat-v3	OpenRouter	39.444%	Details
models · gamedev	DeepSeek-V3-0324deepseek/deepseek-chat-v3	Design Arena	1,066 Elo	Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.

Apps & session costs

Compare observed median session costs for applications using this model, grouped by conversation length.

30-day sample · through Sep 27, 2026

Application	Model route	Turns	Median cost / session	Details
Hermes Agent	deepseek/deepseek-chat-v3	2–9 turns	$0.012295	Details
Claude Code	deepseek/deepseek-chat-v3	1 turn	$0.002275	Details
Hermes Agent	deepseek/deepseek-chat-v3	10–49 turns	$0.105777	Details
Hermes Agent	deepseek/deepseek-chat-v3	1 turn	$0.004862	Details

Source: OpenRouter, as of Oct 2, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications

deepseek/deepseek-chatObserved Oct 4, 2026, 11:31 AM UTC

Context: 163,840
Maximum output: 16,000
Tokenizer: DeepSeek
Input: text
Output: text
Moderation: No
Knowledge cutoff: 2024-07-31
Added to OpenRouter: 2024-12-26

Base URLhttps://openrouter.ai/api/v1

Model IDdeepseek/deepseek-chat

Version IDdeepseek/deepseek-chat-v3

Weights IDdeepseek-ai/DeepSeek-V3

Supported parameters

frequency_penalty
logit_bias
max_tokens
min_p
presence_penalty
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_p

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does DeepSeek Chat cost?

Listed rates start at $0.10 per million input tokens and $0.295 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer DeepSeek Chat?

There are 7 cataloged offers across 6 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 1,000,000 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Yes. Structured output: Not reported. Support can vary by endpoint.

Resources

Hugging Face ↗

Catalog history · 2 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

Sep 14, 2026 · b21a85056ac8
Sep 12, 2026 · b7631a437ba2