LLM 주요 모델 가격 지도 (2026년 3월 기준)

Date: 2026-03-18
Grounding: 공식 가격 페이지에서 직접 추출. 2차 출처와 교차 검증 완료.
목적: REODE Multi LLM Provider 참고 자료로 사용
단위: USD per 1M tokens

---

1. OpenAI

출처: https://www.tldl.io/resources/llm-api-pricing-2026 (3월 갱신)

모델	Input	Output	Cache Input	Context	Status
GPT-5.2 Pro	$21.00	$168.00	$2.10	200K	Current
GPT-5.2	$1.75	$14.00	$0.175	200K	Current
GPT-5	$1.25	$10.00	$0.125	128K	Current
GPT-5 Mini	$0.25	$2.00	$0.025	200K	Current
GPT-5 Nano	$0.05	$0.40	$0.005	128K	Current
o4-mini	$1.10	$4.40	$0.275	200K	Current
o3	$2.00	$8.00	$1.00	200K	Current
o3-pro	$20.00	$80.00	—	200K	Current
GPT-4.1	$2.00	$8.00	$0.20	1M	Current
GPT-4.1 Mini	$0.40	$1.60	$0.04	1M	Current
GPT-4.1 Nano	$0.10	$0.40	$0.01	1M	Current
o1	$15.00	$60.00	$7.50	200K	Maintained
GPT-4o	—	—	—	—	Deprecated → GPT-4.1
GPT-4o mini	—	—	—	—	Deprecated → GPT-4.1 Nano

2. Anthropic

출처: https://www.tldl.io/resources/llm-api-pricing-2026

모델	Input	Output	Cache Input	Context	Status
Claude Opus 4.6	$5.00	$25.00	$0.50	200K	Current
Claude Sonnet 4.6	$3.00	$15.00	$0.30	200K	Current
Claude Haiku 4.5	$1.00	$5.00	$0.10	200K	Current
Claude 3.5 Sonnet	—	—	—	—	Deprecated
Claude 3 Haiku	—	—	—	—	Deprecated

3. Google Gemini

출처: https://www.tldl.io/resources/llm-api-pricing-2026

모델	Input	Output	Cache Input	Context	Status
Gemini 3.1 Pro	$2~4	$12~18	—	200K+	Current
Gemini 3 Flash	$0.50	$3.00	—	—	Current
Gemini 2.5 Pro (≤200K)	$1.25	$10.00	$0.125	2M	Current
Gemini 2.5 Pro (>200K)	$2.50	$15.00	$0.25	2M	Current
Gemini 2.5 Flash	$0.30	$2.50	$0.03	1M	Current
Gemini 2.5 Flash-Lite	$0.10	$0.40	—	1M	Current
Gemini 2.0 Flash	$0.10	$0.40	$0.025	1M	Maintained

4. Zhipu (GLM)

출처: https://docs.z.ai/guides/overview/pricing (공식)

모델	Input	Output	Context	Status
GLM-5	$1.00	$3.20	200K	Current (2026-02)
GLM-5-Turbo	$1.20	$4.00	200K	Current (2026-03-16)
GLM-5-Code	$1.20	$5.00	200K	Current
GLM-4.7	$0.60	$2.20	128K	Maintained
GLM-4.7-FlashX	$0.07	$0.40	128K	Current
GLM-4.7-Flash	Free	Free	128K	Current
GLM-4.6	$0.60	$2.20	128K	Maintained
GLM-4.5	$0.60	$2.20	128K	Maintained
GLM-4.5-X	$2.20	$8.90	128K	Maintained
GLM-4.5-Air	$0.20	$1.10	128K	Maintained
GLM-4.5-AirX	$1.10	$4.50	128K	Maintained
GLM-4.5-Flash	Free	Free	128K	Current
GLM-4-32B-0414-128K	$0.10	$0.10	128K	Maintained
GLM-4	—	—	—	Legacy
GLM-3	—	—	—	Deprecated

Cache: 전 모델 cached input 80% 할인. Storage 한시적 무료.

5. MiniMax

출처: https://pricepertoken.com/pricing-page/provider/minimax (교차 검증)

모델	Input	Output	Context	Status	출시
MiniMax-M2.5	$0.25	$0.95	197K	Current	2026-02
MiniMax-M2 Her	$0.30	$1.20	66K	Current	2026-01
MiniMax-M2.1	$0.27	$0.95	197K	Maintained	2025-12
MiniMax-M2	$0.255	$1.00	197K	Maintained	2025-10
MiniMax-M1	$0.40	$1.76	1M	Legacy	2025-06
MiniMax-01	$0.20	$1.10	1M	Maintained	—
abab6.5	—	—	—	Deprecated

6. DeepSeek

출처: https://api-docs.deepseek.com/quick_start/pricing (공식)

모델	Input	Output	Cache Input	Context	Status
deepseek-chat (V3.2 non-thinking)	$0.28	$0.42	$0.028	128K	Current
deepseek-reasoner (V3.2 thinking)	$0.28	$0.42	$0.028	128K	Current
DeepSeek V3.2 Speciale	$0.40	$1.20	—	128K	Current
DeepSeek V3.1	—	—	—	—	→ V3.2 통합
DeepSeek V3	—	—	—	—	→ V3.2 통합
DeepSeek R1	—	—	—	—	→ V3.2 reasoning 통합

7. Qwen (Alibaba)

출처: https://pricepertoken.com/pricing-page/provider/qwen (교차 검증)

모델	Input	Output	Context	Status
Qwen3.5-397B-A17B	$0.39	$0.90	262K	Current (2026-02)
Qwen3.5-122B-A10B	$0.26	$2.08	262K	Current
Qwen3.5 Plus	$0.26	$1.56	1M	Current
Qwen3.5-Flash	$0.07	$0.26	1M	Current
Qwen3.5-35B-A3B	$0.163	$1.00	262K	Current
Qwen3.5-27B	$0.195	$1.56	262K	Current
Qwen3.5-9B	$0.05	$0.15	262K	Current
Qwen3 Max	$0.78	$3.90	262K	Current
Qwen3-235B-A22B	$0.455	$1.82	—	Current
Qwen3-30B-A3B	$0.08	$0.28	41K	Current
Qwen3 32B	$0.08	$0.24	41K	Current
Qwen3 14B	$0.06	$0.20	41K	Current
Qwen3 8B	$0.05	$0.20	41K	Current
Qwen-Plus	$0.26	$0.78	1M	Maintained
Qwen-Max (legacy)	$1.04	$4.16	33K	Legacy

Qwen3 Coder 계열 (pricepertoken 기준):

모델	Input	Output	Context	Status
Qwen3 Coder (480B-A35B)	$0.22	$0.90	262K	Current
Qwen3 Coder-Next	$0.50	$1.20	262K	Current

8. Kimi (Moonshot)

출처: https://costgoat.com/pricing/kimi-api (교차 검증)

모델	Input	Output	Context	Status
Kimi K2.5	$0.60	$3.00	262K	Current (2026-01)
Kimi K2 Thinking	$0.60	$2.50	262K	Current
Kimi K2 0905	$0.60	$2.50	262K	Maintained
Kimi K2 0711	$0.60	$2.50	131K	Maintained
Kimi K2 Turbo	$1.15	$8.00	262K	Current
Kimi K2 Thinking Turbo	$1.15	$8.00	262K	Current
Moonshot V1 8K	$0.20	$2.00	8K	Legacy
Moonshot V1 32K	$1.00	$3.00	32K	Legacy
Moonshot V1 128K	$2.00	$5.00	131K	Legacy

Cache: $0.15/M (75% 절감). Web search $0.005/call.

9. Mistral (코딩 모델 포함)

출처: https://pricepertoken.com/pricing-page/provider/mistral-ai (2026-03-17 갱신)

 

모델	Input	Output	Context	Status
Devstral Small 2505	Free	Free	128K	Current
Mistral Nemo	$0.02	$0.04	131K	Current
Ministral 3B	$0.04	$0.04	128K	Current
Mistral Small 3.2 24B	$0.06	$0.18	131K	Current
Devstral Small 1.1	$0.07	$0.28	131K	Current
Mistral Small 3.1 24B	$0.10	$0.30	128K	Current
Ministral 8B	$0.10	$0.10	128K	Current
Codestral 2508	$0.30	$0.90	256K	Current
Devstral 2 2512	$0.40	$0.90	262K	Current
Devstral Medium	$0.40	$2.00	131K	Current
Mistral Medium 3.1	$0.40	$2.00	131K	Current
Mistral Large 3 2512	$0.50	$1.50	262K	Current
Magistral Medium	$2.00	$5.00	40K	Current

10. xAI (Grok)

출처: https://www.tldl.io/resources/llm-api-pricing-2026

모델	Input	Output	Context	Status
Grok 4	$3.00	$15.00	2M	Current
Grok 4.1 Fast	$0.20	$0.50	2M	Current

---

11. 코스트 효율 랭킹 (Output ≤$2/M, Tool Calling 지원)

Output 가격 기준 오름차순:

#	모델	Provider	Output $/M	Input $/M	Context	특기
1	GLM-4-32B-0414	Zhipu	$0.10	$0.10	128K	최저가 실용급
2	Qwen3.5-9B	Alibaba	$0.15	$0.05	262K	초경량, 262K
3	Mistral Small 3.2	Mistral	$0.18	$0.06	131K	유럽 서버
4	Qwen3 8B	Alibaba	$0.20	$0.05	41K	초경량
5	Qwen3.5-Flash	Alibaba	$0.26	$0.07	1M	1M context
6	Qwen3-30B-A3B	Alibaba	$0.28	$0.08	41K	MoE 경량
7	Devstral Small 1.1	Mistral	$0.28	$0.07	131K	코딩 특화
8	Mistral Small 3.1	Mistral	$0.30	$0.10	128K	범용
9	GPT-5 Nano	OpenAI	$0.40	$0.05	128K	OpenAI 최저가
10	GPT-4.1 Nano	OpenAI	$0.40	$0.10	1M	1M context
11	Gemini 2.5 Flash-Lite	Google	$0.40	$0.10	1M	1M context
12	GLM-4.7-FlashX	Zhipu	$0.40	$0.07	128K	GLM 고속
13	DeepSeek V3.2	DeepSeek	$0.42	$0.28	128K	가성비 왕, chat+reasoning
14	Grok 4.1 Fast	xAI	$0.50	$0.20	2M	2M context 최대
15	Qwen-Plus	Alibaba	$0.78	$0.26	1M	범용
16	Qwen3 Coder	Alibaba	$0.90	$0.22	262K	코딩 전용 MoE
17	Devstral 2	Mistral	$0.90	$0.40	262K	SWE-bench 72.2%
18	MiniMax M2.5	MiniMax	$0.95	$0.25	197K	에이전트 호환
19	MiniMax-01	MiniMax	$1.10	$0.20	1M	1M context
20	Mistral Large 3	Mistral	$1.50	$0.50	262K	범용 대형
21	Qwen3.5 Plus	Alibaba	$1.56	$0.26	1M	1M, 최신
22	MiniMax M1	MiniMax	$1.76	$0.40	1M	Legacy
23	Kimi K2 0905	Moonshot	$1.90	$0.39	262K	200+ tool call 안정

무료 모델

모델	Provider	Context	Status
GLM-4.7-Flash	Zhipu	128K	Current
GLM-4.5-Flash	Zhipu	128K	Current
Devstral Small 2505	Mistral	128K	Current
Llama 4	Meta (via providers)	200K	Current

---

12. REODE 용도별 추천

용도	추천 모델	Output $/M	이유
메인 에이전트	Claude Opus 4.6	$25.00	최고 성능, 기존 통합
코딩 서브에이전트	Qwen3 Coder ($0.90) 또는 Devstral 2 ($0.90)	$0.90	코딩 특화, tool calling
경량 서브에이전트	DeepSeek V3.2 ($0.42)	$0.42	최저가 실용급
대량 병렬 호출	Qwen3.5-9B ($0.15) 또는 GLM-4-32B ($0.10)	$0.10~0.15	초저가
장문 컨텍스트	Grok 4.1 Fast ($0.50, 2M) 또는 GPT-4.1 Nano ($0.40, 1M)	$0.40~0.50	1M+ context
무료 테스트	GLM-4.7-Flash	Free	개발/테스트용

---

Sources (공식 1차 소스)

• Z.AI Pricing (GLM) — 공식 가격 페이지에서 직접 추출
• DeepSeek Pricing — 공식 API 문서에서 직접 추출
• MiniMax Pricing — pricepertoken.com 교차 검증
• Qwen Pricing — pricepertoken.com + alibabacloud.com 교차 검증
• Kimi Pricing — platform.moonshot.ai 교차 검증
• Mistral Pricing — 2026-03-17 갱신 확인
• TLDL LLM Pricing March 2026 — OpenAI/Anthropic/Google/xAI
• pricepertoken.com Cheapest — 전체 비교
• Awesome Agents Pricing — 교차 검증