#가격 #비교: LLM API 서비스 - 모델 통합 라우터

BIBLIOGRAPHY

“Models & Pricing | Deepseek Api Docs.” n.d. Accessed March 21, 2025. https://api-docs.deepseek.com/quick_start/pricing/.

“Pricing Perplexity Sonar 가격.” n.d. Perplexity. Accessed March 21, 2025. https://perplexity.mintlify.app/guides/pricing.

History

[2025-01-27 Mon 16:18] 관심 가지고 만들었음
[2025-03-21 Fri 10:19] #비교, #결제 #가격 #비용 #청구 서비스 호출

2025 API 서비스 가격 비교

[2025-01-27 Mon 16:18]

LLM Pricing 가격비교

나름 정리해본다면?

updated [2025-01-27 Mon 20:59]

Provider	Model	Input/M tokens	Output/M tokens	Modalities	context	Notes	ref
Anthropic	claude-haiku	0.25	1.25				Anthropic Console
	claude-sonnet	3.00	15.00
	claude-opus	15.00	75.00			$yikes
	claude-3.5-sonnet
OpenAI	gpt-4-0125-preview	10.00	30.00				Pricing
	gpt-4o-mini	0.15	0.60	text,vision	128K
Together.ai	41B-70B chat models	0.90 (in+out)					Together Pricing
deepseek	deepseek-chat	0.07 0.014	1.1 0.28		64K		Models & Pricing
	deepseek-reasoner	0.14	2.19		64K
perplexity	sonar	1	1		127K
	sonar-pro	3	15		200K

perplexity - 추론

https://perplexity.mintlify.app/guides/pricing (“Pricing Perplexity Sonar 가격” n.d.)

sonar-reasoning 127k Chat Completion
- sonar-reasoning outputs CoT in its response as well
sonar-pro 200k Chat Completion
- sonar-pro has a max output token limit of 8k
sonar 127k Chat Completion

upstage

10달러 쿠폰

solar-pro, 22B, 32k, $0.25 / 1M tokens
solar-mini, 10.7b, 32k, $0.15 / 1M tokens
Console - Upstage - console.upstage.ai

deepseek

(“Models & Pricing | Deepseek Api Docs” n.d.) [2025-01-28 Tue 12:08]

openrouter

DeepSeek V3 - API, Providers, Stats | OpenRouter - openrouter.ai
- 64k context $0.14/ M in p u tt o k e n s$ 0.28/M output tokens
Gemini Flash 1.5 - API, Providers, Stats | OpenRouter - openrouter.ai
- 100k context $0.075/ M in p u tt o k e n s$ 0.3/M output tokens
Claude 3.5 Sonnet - API, Providers, Stats | OpenRouter - openrouter.ai
- 200k context $3/ M in p u tt o k e n s$ 15/M output tokens
GPT-4o-mini - API, Providers, Stats | OpenRouter - openrouter.ai
- 128k context $0.15/ M in p u tt o k e n s$ 0.6/M output tokens
anthropic/claude-3.5-haiku
- 200,000 context $0.8/ M in p u tt o k e n s$ 4/M output tokens

gptel model guide - 가격 정보 넣는 방법

;;;; model descriptions
 
;; updated 2025-01-27
 
;; sonar, sonar-pro
 
;; (defconst gptel--perplexity-models
;;   '((sonar
;;      :description "Google's latest model with enhanced capabilities across various tasks"
;;      :capabilities (tool-use json media)
;;      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
;;                   "application/pdf" "text/plain" "text/csv" "text/html")
;;      :context-window 2000
;;      ;; input & output price is halved for prompts of 128k tokens or less
;;      :input-cost 2.50
;;      :output-cost 10
;;      :cutoff-date "2024-05")
;;     (gemini-2.0-flash-exp
;;      :description "Next generation features, superior speed, native tool use"
;;      :capabilities (tool-use json media)
;;      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
;;                   "application/pdf" "text/plain" "text/csv" "text/html")
;;      :context-window 1000
;;      :cutoff-date "2024-12")
;;     (gemini-1.5-flash
;;      :description "A faster, more efficient version of Gemini 1.5 optimized for speed"
;;      :capabilities (tool-use json media)
;;      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
;;                   "application/pdf" "text/plain" "text/csv" "text/html")
;;      :context-window 1000
;;      ;; input & output price is halved for prompts of 128k tokens or less
;;      :input-cost 0.15
;;      :output-cost 0.60
;;      :cutoff-date "2024-05")
;;     (gemini-1.5-flash-8b
;;      :description "High volume and lower intelligence tasks"
;;      :capabilities (tool-use json media)
;;      :context-window 1000
;;      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
;;                   "application/pdf" "text/plain" "text/csv" "text/html")
;;      ;; input & output price is halved for prompts of 128k tokens or less
;;      :input-cost 0.075
;;      :output-cost 0.30
;;      :cutoff-date "2024-10")
;;     (gemini-2.0-flash-thinking-exp
;;      :description "Stronger reasoning capabilities."
;;      :capabilities (tool-use media)
;;      :context-window 32
;;      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
;;                   "text/plain" "text/csv" "text/html")
;;      :cutoff-date "2024-08")
;;     (gemini-exp-1206
;;      :description "Improved coding, reasoning and vision capabilities"
;;      :capabilities (tool-use json media)
;;      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
;;                   "application/pdf" "text/plain" "text/csv" "text/html")
;;      :cutoff-date "2024-12")
;;     (gemini-pro
;;      :description "The previous generation of Google's multimodal AI model"
;;      :capabilities (tool-use json media)
;;      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
;;                   "application/pdf" "text/plain" "text/csv" "text/html")
;;      :context-window 32
;;      :input-cost 0.50
;;      :output-cost 1.50
;;      :cutoff-date "2023-02"))
;;   "List of available Gemini models and associated properties.
;; Keys:
 
;; - `:description': a brief description of the model.
 
;; - `:capabilities': a list of capabilities supported by the model.
 
;; - `:mime-types': a list of supported MIME types for media files.
 
;; - `:context-window': the context window size, in thousands of tokens.
 
;; - `:input-cost': the input cost, in US dollars per million tokens.
 
;; - `:output-cost': the output cost, in US dollars per million tokens.
 
;; - `:cutoff-date': the knowledge cutoff date.
 
;; - `:request-params': a plist of additional request parameters to
;;   include when using this model.
 
;; Information about the Gemini models was obtained from the following
;; source:
 
;; - <https://ai.google.dev/pricing>
;; - <https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models>")

🧠 Junghanacs

Table of Contents