LLM Pricing
PricingLeaderboardsToolsProvidersReleasesGuides

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

DeepSeek Reasoner

deepseek·deepseek/deepseek-reasoner·GA·Open weights·deepseek-thinking series
DeepSeek reasoning model for multi-step analysis, math, coding, and tools

DeepSeek Reasoner is offered by 4 providers on this page. Public prices are shown for 3 of them. Its reference price is $0.56 per 1M input tokens and $1.68 per 1M output tokens. The lowest paid channel is OrcaRouter at $0.147 / $0.295 per 1M, about 3.8× below the reference price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it.

The context window is 128,000 tokens at the reference host, but hosts report different limits, from 64,000 to 1,000,000, so the usable window depends on the provider you pick. It supports reasoning and tool use. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2025-09.

Specs & pricing

Input / output per 1M tokens
Reference price·Helicone
$0.56 / $1.68
Blended $0.84 · Cache read $0.07
Lowest paid·OrcaRouterGateway
$0.15 / $0.30
Blended $0.18 · 3.8× spread
Context
128,000
Output limit
64,000
Knowledge cutoff
2025-09
Released / updated
2025-12-01 / 2026-02-28
Capabilities
✓ Reasoning✓ Tool useStructured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
Text
Weights
Hugging Face

Available at 4 providers3 with public prices

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
OrcaRouterGateway$0.15$0.30$0.028—1,000,000 ⚠384,000
NanoGPTGateway$0.40$1.70$0.20—64,000 ⚠—
HeliconeGateway$0.56$1.68$0.07—128,00064,000
AnyAPI
deepseek/deepseek-r1
Gateway————1,000,000 ⚠384,000hostTable.undisclosed

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

Upstream provides no control info

Interleaved thinking (reasoning between tool calls) is declared by 2 of 4 providers.

2 / 4 providers expose no reasoning control (reasoning_options: []).

Your usage cost

1OrcaRouter$29.87
2Helicone$137.20
3NanoGPT$141.00
The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$1.68$02026-08-052026-08-282026-08-05 · Input list $0.142026-08-06 · Input list $0.142026-08-07 · Input list $0.142026-08-08 · Input list $0.142026-08-09 · Input list $0.142026-08-10 · Input list $0.142026-08-11 · Input list $0.142026-08-12 · Input list $0.142026-08-13 · Input list $0.142026-08-26 · Input list $0.562026-08-28 · Input list $0.562026-08-05 · Output list $0.282026-08-06 · Output list $0.282026-08-07 · Output list $0.282026-08-08 · Output list $0.282026-08-09 · Output list $0.282026-08-10 · Output list $0.282026-08-11 · Output list $0.282026-08-12 · Output list $0.282026-08-13 · Output list $0.282026-08-26 · Output list $1.682026-08-28 · Output list $1.682026-08-05 · Min blended $0.182026-08-06 · Min blended $0.182026-08-07 · Min blended $0.182026-08-08 · Min blended $0.182026-08-09 · Min blended $0.182026-08-10 · Min blended $0.182026-08-11 · Min blended $0.182026-08-12 · Min blended $0.182026-08-13 · Min blended $0.182026-08-26 · Min blended $0.332026-08-28 · Min blended $0.18

Benchmark1 items

NameConditionsScoreMetricSource
Aider Polyglot—74.2percent correctSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Related models

DeepSeek V4 Pro 0813 (Databricks)same series$1.32 / $3.96DeepSeek V4 Pro 0813 (EU)same series$1.75 / $3.50DeepSeek V4 Pro 0813 Thinkingsame series$1.10 / $2.50GLM-5.3-Flashcheaper alternative$0.15 / $0.50DeepSeek V4.1 Flashcheaper alternative$0.15 / $0.60GPT-5.6 Lunacheaper alternative$0.20 / $1.20GPT-5.4 nanocheaper alternative$0.20 / $1.25
Data partly from models.dev (MIT) · Helicone official docs ↗