model cost comparison

Estimated monthly cost for AI Agent workflows. Costs are calculated based on typical agent task patterns — multi-turn tool-calling chains with context accumulation.

Regular — Daily use: ~10 agent tasks/day, ~100 LLM calls per task. Typical for a professional running multiple daily workflows.
300Tasks / Month
30,000LLM Calls
90.0MInput Tokens
15.0MOutput Tokens

64 models from 22 providers

ModelProviderContextFeaturesMonthly Cost ↑Calculation
MiniMax M2.7
MiniMax-M2.7		MiniMax Token Plan (China)200K$4.20Starter plan Y29/mo (~36,000 requests/mo based on 2x5h/day). 30000 calls needed.
MiniMax M2.7
MiniMax-M2.7		MiniMax Token Plan (International)200K$10.00Starter plan $10/mo (~90,000 requests/mo based on 2x5h/day). 30000 calls needed.
Qwen3.5 Plus
qwen3.5-plus		Alibaba Cloud1.0M$20.8790.0M x ¥0.80/M + 15.00M x ¥4.80/M = ¥72.00 + ¥72.00
Mistral Small 4
mistral-small-latest		Mistral131K$22.5090.0M x $0.15/M + 15.00M x $0.60/M = $13.50 + $9.00
Grok 4-1 Fast (Reasoning)
grok-4-1-fast-reasoning		xAI2.0M$25.5090.0M x $0.20/M + 15.00M x $0.50/M = $18.00 + $7.50
Grok 4-1 Fast (Non-Reasoning)
grok-4-1-fast-non-reasoning		xAI2.0M$25.5090.0M x $0.20/M + 15.00M x $0.50/M = $18.00 + $7.50
MiniMax M2.5
MiniMax-M2.5		Volcengine (Coding Plan)200K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 4.7
GLM-4.7		Volcengine (Coding Plan)200K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Kimi-K2.5		Volcengine (Coding Plan)262K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Doubao Seed 2.0 Pro
Doubao-Seed-2.0-pro		Volcengine (Coding Plan)256K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Doubao Seed 2.0 Code
Doubao-Seed-2.0-Code		Volcengine (Coding Plan)256K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
DeepSeek V3.2
DeepSeek-V3.2		Volcengine (Coding Plan)128K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
MiniMax M2.5
MiniMax-M2.5		Baidu Qianfan (Coding Plan)200K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 5
GLM-5		Baidu Qianfan (Coding Plan)128K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 4.7
GLM-4.7		Baidu Qianfan (Coding Plan)200K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Kimi-K2.5		Baidu Qianfan (Coding Plan)262K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
DeepSeek V3.2
DeepSeek-V3.2		Baidu Qianfan (Coding Plan)128K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
MiniMax M2.5
MiniMax-M2.5		Alicloud (Coding Plan)200K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 4.7
GLM-4.7		Alicloud (Coding Plan)200K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 5
GLM-5		Alicloud (Coding Plan)128K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Kimi-K2.5		Alicloud (Coding Plan)262K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Qwen 3.5 Plus
Qwen3.5-Plus		Alicloud (Coding Plan)1.0M$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Qwen 3 Max
Qwen3-Max-2026-01-23		Alicloud (Coding Plan)262K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
MiniMax M2.5
MiniMax-M2.5		Tencent Cloud (Coding Plan)200K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 5
GLM-5		Tencent Cloud (Coding Plan)128K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Kimi-K2.5		Tencent Cloud (Coding Plan)262K$28.99Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
DeepSeek V3.2 (Chat)
deepseek-chat		DeepSeek131K$31.5090.0M x $0.28/M + 15.00M x $0.42/M = $25.20 + $6.30
DeepSeek V3.2 (Reasoner)
deepseek-reasoner		DeepSeek131K$31.5090.0M x $0.28/M + 15.00M x $0.42/M = $25.20 + $6.30
DeepSeek V3.2 (via Volcengine)
deepseek-v3-2-251201		Volcengine128K$32.6190.0M x Y2.00/M + 15.00M x Y3.00/M = Y180.00 + Y45.00
DeepSeek V3.2 (via Qianfan)
deepseek-v3.2		Baidu Qianfan98K$32.6190.0M x Y2/M + 15.00M x Y3/M = Y180.00 + Y45.00
MiniMax M2.5 (via OpenRouter)
minimax/minimax-m2.5		OpenRouter200K$35.5590.0M x $0.20/M + 15.00M x $1.17/M = $18.00 + $17.55
DeepSeek V3.2 (via OpenRouter)
deepseek/deepseek-chat		OpenRouter131K$42.1590.0M x $0.32/M + 15.00M x $0.89/M = $28.80 + $13.35
GLM-4.7
glm-4.7		Zhipu AI (China)200K$43.4890.0M x ¥2.00/M + 15.00M x ¥8.00/M = ¥180.00 + ¥120.00
MiniMax M2.7
MiniMax-M2.7		MiniMax200K$45.0090.0M x $0.30/M + 15.00M x $1.20/M = $27.00 + $18.00
MiniMax M2.5
MiniMax-M2.5		MiniMax200K$45.0090.0M x $0.30/M + 15.00M x $1.20/M = $27.00 + $18.00
MiniMax M2.7
MiniMax-M2.7		MiniMax (International)200K$45.0090.0M x $0.30/M + 15.00M x $1.20/M = $27.00 + $18.00
GLM 4.7
zai-org/GLM-4.7		Together AI203K$47.2590.0M x $0.45/M + 15.00M x $0.45/M = $40.50 + $6.75
Qwen3 Max
qwen3-max		Alibaba Cloud262K$54.3590.0M x ¥2.50/M + 15.00M x ¥10.00/M = ¥225.00 + ¥150.00
GLM 4.7 (via OpenRouter)
z-ai/glm-4.7		OpenRouter200K$61.3590.0M x $0.39/M + 15.00M x $1.75/M = $35.10 + $26.25
Gemini 2.5 Flash
gemini-2.5-flash		Google1.0M$64.5090.0M x $0.30/M + 15.00M x $2.50/M = $27.00 + $37.50
Devstral 2
devstral-latest		Mistral131K$66.0090.0M x $0.40/M + 15.00M x $2.00/M = $36.00 + $30.00
Mistral Large 3
mistral-large-latest		Mistral131K$67.5090.0M x $0.50/M + 15.00M x $1.50/M = $45.00 + $22.50
Qwen 3.5 (via OpenRouter)
qwen/qwen3.5-397b-a17b		OpenRouter262K$70.2090.0M x $0.39/M + 15.00M x $2.34/M = $35.10 + $35.10
Kimi K2.5 (via OpenRouter)
moonshotai/kimi-k2.5		OpenRouter262K$73.5090.0M x $0.45/M + 15.00M x $2.20/M = $40.50 + $33.00
Kimi K2.5 (via Volcengine)
kimi-k2-5-260127		Volcengine256K$86.9690.0M x Y4.00/M + 15.00M x Y16.00/M = Y360.00 + Y240.00
Kimi K2.5 (via Together)
moonshotai/Kimi-K2.5		Together AI262K$87.0090.0M x $0.50/M + 15.00M x $2.80/M = $45.00 + $42.00
Gemini 3 Flash Preview
gemini-3-flash-preview		Google1.0M$90.0090.0M x $0.50/M + 15.00M x $3.00/M = $45.00 + $45.00
GLM-5
glm-5		Zhipu AI (China)128K$91.3090.0M x ¥4.00/M + 15.00M x ¥18.00/M = ¥360.00 + ¥270.00
Kimi K2.5
kimi-k2.5		Moonshot (International)262K$99.0090.0M x $0.60/M + 15.00M x $3.00/M = $54.00 + $45.00
GLM 5 (via OpenRouter)
z-ai/glm-5		OpenRouter128K$99.3090.0M x $0.72/M + 15.00M x $2.30/M = $64.80 + $34.50
DeepSeek R1 (via OpenRouter)
deepseek/deepseek-r1		OpenRouter164K$100.5090.0M x $0.70/M + 15.00M x $2.50/M = $63.00 + $37.50
GLM-5-Turbo
glm-5-turbo		Zhipu AI (China)128K$113.0490.0M x ¥5.00/M + 15.00M x ¥22.00/M = ¥450.00 + ¥330.00
Qwen 3 Max (via OpenRouter)
qwen/qwen3-max		OpenRouter262K$128.7090.0M x $0.78/M + 15.00M x $3.90/M = $70.20 + $58.50
GPT-5.4 mini
gpt-5.4-mini		OpenAI1.1M$135.0090.0M x $0.75/M + 15.00M x $4.50/M = $67.50 + $67.50
Kimi K2.5
kimi-k2.5		Moonshot (China)262K$144.0090.0M x $1.10/M + 15.00M x $3.00/M = $99.00 + $45.00
Gemini 2.5 Pro
gemini-2.5-pro		Google1.0M$262.5090.0M x $1.25/M + 15.00M x $10.00/M = $112.50 + $150.00
Grok 4.20 Beta (Reasoning)
grok-4.20-0309-reasoning		xAI2.0M$270.0090.0M x $2.00/M + 15.00M x $6.00/M = $180.00 + $90.00
Grok 4.20 Beta (Non-Reasoning)
grok-4.20-0309-non-reasoning		xAI2.0M$270.0090.0M x $2.00/M + 15.00M x $6.00/M = $180.00 + $90.00
Gemini 3.1 Pro Preview
gemini-3.1-pro-preview		Google1.0M$360.0090.0M x $2.00/M + 15.00M x $12.00/M = $180.00 + $180.00
GPT-5.3 Codex
gpt-5.3-codex		OpenAI1.1M$367.5090.0M x $1.75/M + 15.00M x $14.00/M = $157.50 + $210.00
DeepSeek R1
deepseek-ai/DeepSeek-R1		Together AI164K$375.0090.0M x $3.00/M + 15.00M x $7.00/M = $270.00 + $105.00
GPT-5.4
gpt-5.4		OpenAI1.1M$450.0090.0M x $2.50/M + 15.00M x $15.00/M = $225.00 + $225.00
Claude Sonnet 4.6
claude-sonnet-4-6		Anthropic1.0M$495.0090.0M x $3.00/M + 15.00M x $15.00/M = $270.00 + $225.00
Claude Opus 4.6
claude-opus-4-6		Anthropic1.0M$825.0090.0M x $5.00/M + 15.00M x $25.00/M = $450.00 + $375.00
MiniMax M2.7
MiniMax Token Plan (China)
$4.20/month
ctx 200K
Starter plan Y29/mo (~36,000 requests/mo based on 2x5h/day). 30000 calls needed.
MiniMax M2.7
MiniMax Token Plan (International)
$10.00/month
ctx 200K
Starter plan $10/mo (~90,000 requests/mo based on 2x5h/day). 30000 calls needed.
Qwen3.5 Plus
Alibaba Cloud
$20.87/month
ctx 1.0M
90.0M x ¥0.80/M + 15.00M x ¥4.80/M = ¥72.00 + ¥72.00
Mistral Small 4
Mistral
$22.50/month
ctx 131K
90.0M x $0.15/M + 15.00M x $0.60/M = $13.50 + $9.00
Grok 4-1 Fast (Reasoning)
xAI
$25.50/month
ctx 2.0M
90.0M x $0.20/M + 15.00M x $0.50/M = $18.00 + $7.50
Grok 4-1 Fast (Non-Reasoning)
xAI
$25.50/month
ctx 2.0M
90.0M x $0.20/M + 15.00M x $0.50/M = $18.00 + $7.50
MiniMax M2.5
Volcengine (Coding Plan)
$28.99/month
ctx 200K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 4.7
Volcengine (Coding Plan)
$28.99/month
ctx 200K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Volcengine (Coding Plan)
$28.99/month
ctx 262K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Doubao Seed 2.0 Pro
Volcengine (Coding Plan)
$28.99/month
ctx 256K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Doubao Seed 2.0 Code
Volcengine (Coding Plan)
$28.99/month
ctx 256K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
DeepSeek V3.2
Volcengine (Coding Plan)
$28.99/month
ctx 128K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
MiniMax M2.5
Baidu Qianfan (Coding Plan)
$28.99/month
ctx 200K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 5
Baidu Qianfan (Coding Plan)
$28.99/month
ctx 128K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 4.7
Baidu Qianfan (Coding Plan)
$28.99/month
ctx 200K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Baidu Qianfan (Coding Plan)
$28.99/month
ctx 262K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
DeepSeek V3.2
Baidu Qianfan (Coding Plan)
$28.99/month
ctx 128K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
MiniMax M2.5
Alicloud (Coding Plan)
$28.99/month
ctx 200K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 4.7
Alicloud (Coding Plan)
$28.99/month
ctx 200K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 5
Alicloud (Coding Plan)
$28.99/month
ctx 128K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Alicloud (Coding Plan)
$28.99/month
ctx 262K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Qwen 3.5 Plus
Alicloud (Coding Plan)
$28.99/month
ctx 1.0M
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Qwen 3 Max
Alicloud (Coding Plan)
$28.99/month
ctx 262K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
MiniMax M2.5
Tencent Cloud (Coding Plan)
$28.99/month
ctx 200K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
GLM 5
Tencent Cloud (Coding Plan)
$28.99/month
ctx 128K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
Kimi K2.5
Tencent Cloud (Coding Plan)
$28.99/month
ctx 262K
Pro plan Y200/mo (~90,000 requests/mo). 30000 calls needed.
DeepSeek V3.2 (Chat)
DeepSeek
$31.50/month
ctx 131K
90.0M x $0.28/M + 15.00M x $0.42/M = $25.20 + $6.30
DeepSeek V3.2 (Reasoner)
DeepSeek
$31.50/month
ctx 131K
90.0M x $0.28/M + 15.00M x $0.42/M = $25.20 + $6.30
DeepSeek V3.2 (via Volcengine)
Volcengine
$32.61/month
ctx 128K
90.0M x Y2.00/M + 15.00M x Y3.00/M = Y180.00 + Y45.00
DeepSeek V3.2 (via Qianfan)
Baidu Qianfan
$32.61/month
ctx 98K
90.0M x Y2/M + 15.00M x Y3/M = Y180.00 + Y45.00
MiniMax M2.5 (via OpenRouter)
OpenRouter
$35.55/month
ctx 200K
90.0M x $0.20/M + 15.00M x $1.17/M = $18.00 + $17.55
DeepSeek V3.2 (via OpenRouter)
OpenRouter
$42.15/month
ctx 131K
90.0M x $0.32/M + 15.00M x $0.89/M = $28.80 + $13.35
GLM-4.7
Zhipu AI (China)
$43.48/month
ctx 200K
90.0M x ¥2.00/M + 15.00M x ¥8.00/M = ¥180.00 + ¥120.00
MiniMax M2.7
MiniMax
$45.00/month
ctx 200K
90.0M x $0.30/M + 15.00M x $1.20/M = $27.00 + $18.00
MiniMax M2.5
MiniMax
$45.00/month
ctx 200K
90.0M x $0.30/M + 15.00M x $1.20/M = $27.00 + $18.00
MiniMax M2.7
MiniMax (International)
$45.00/month
ctx 200K
90.0M x $0.30/M + 15.00M x $1.20/M = $27.00 + $18.00
GLM 4.7
Together AI
$47.25/month
ctx 203K
90.0M x $0.45/M + 15.00M x $0.45/M = $40.50 + $6.75
Qwen3 Max
Alibaba Cloud
$54.35/month
ctx 262K
90.0M x ¥2.50/M + 15.00M x ¥10.00/M = ¥225.00 + ¥150.00
GLM 4.7 (via OpenRouter)
OpenRouter
$61.35/month
ctx 200K
90.0M x $0.39/M + 15.00M x $1.75/M = $35.10 + $26.25
Gemini 2.5 Flash
Google
$64.50/month
ctx 1.0M
90.0M x $0.30/M + 15.00M x $2.50/M = $27.00 + $37.50
Devstral 2
Mistral
$66.00/month
ctx 131K
90.0M x $0.40/M + 15.00M x $2.00/M = $36.00 + $30.00
Mistral Large 3
Mistral
$67.50/month
ctx 131K
90.0M x $0.50/M + 15.00M x $1.50/M = $45.00 + $22.50
Qwen 3.5 (via OpenRouter)
OpenRouter
$70.20/month
ctx 262K
90.0M x $0.39/M + 15.00M x $2.34/M = $35.10 + $35.10
Kimi K2.5 (via OpenRouter)
OpenRouter
$73.50/month
ctx 262K
90.0M x $0.45/M + 15.00M x $2.20/M = $40.50 + $33.00
Kimi K2.5 (via Volcengine)
Volcengine
$86.96/month
ctx 256K
90.0M x Y4.00/M + 15.00M x Y16.00/M = Y360.00 + Y240.00
Kimi K2.5 (via Together)
Together AI
$87.00/month
ctx 262K
90.0M x $0.50/M + 15.00M x $2.80/M = $45.00 + $42.00
Gemini 3 Flash Preview
Google
$90.00/month
ctx 1.0M
90.0M x $0.50/M + 15.00M x $3.00/M = $45.00 + $45.00
GLM-5
Zhipu AI (China)
$91.30/month
ctx 128K
90.0M x ¥4.00/M + 15.00M x ¥18.00/M = ¥360.00 + ¥270.00
Kimi K2.5
Moonshot (International)
$99.00/month
ctx 262K
90.0M x $0.60/M + 15.00M x $3.00/M = $54.00 + $45.00
GLM 5 (via OpenRouter)
OpenRouter
$99.30/month
ctx 128K
90.0M x $0.72/M + 15.00M x $2.30/M = $64.80 + $34.50
DeepSeek R1 (via OpenRouter)
OpenRouter
$100.50/month
ctx 164K
90.0M x $0.70/M + 15.00M x $2.50/M = $63.00 + $37.50
GLM-5-Turbo
Zhipu AI (China)
$113.04/month
ctx 128K
90.0M x ¥5.00/M + 15.00M x ¥22.00/M = ¥450.00 + ¥330.00
Qwen 3 Max (via OpenRouter)
OpenRouter
$128.70/month
ctx 262K
90.0M x $0.78/M + 15.00M x $3.90/M = $70.20 + $58.50
GPT-5.4 mini
OpenAI
$135.00/month
ctx 1.1M
90.0M x $0.75/M + 15.00M x $4.50/M = $67.50 + $67.50
Kimi K2.5
Moonshot (China)
$144.00/month
ctx 262K
90.0M x $1.10/M + 15.00M x $3.00/M = $99.00 + $45.00
Gemini 2.5 Pro
Google
$262.50/month
ctx 1.0M
90.0M x $1.25/M + 15.00M x $10.00/M = $112.50 + $150.00
Grok 4.20 Beta (Reasoning)
xAI
$270.00/month
ctx 2.0M
90.0M x $2.00/M + 15.00M x $6.00/M = $180.00 + $90.00
Grok 4.20 Beta (Non-Reasoning)
xAI
$270.00/month
ctx 2.0M
90.0M x $2.00/M + 15.00M x $6.00/M = $180.00 + $90.00
Gemini 3.1 Pro Preview
Google
$360.00/month
ctx 1.0M
90.0M x $2.00/M + 15.00M x $12.00/M = $180.00 + $180.00
GPT-5.3 Codex
OpenAI
$367.50/month
ctx 1.1M
90.0M x $1.75/M + 15.00M x $14.00/M = $157.50 + $210.00
DeepSeek R1
Together AI
$375.00/month
ctx 164K
90.0M x $3.00/M + 15.00M x $7.00/M = $270.00 + $105.00
GPT-5.4
OpenAI
$450.00/month
ctx 1.1M
90.0M x $2.50/M + 15.00M x $15.00/M = $225.00 + $225.00
Claude Sonnet 4.6
Anthropic
$495.00/month
ctx 1.0M
90.0M x $3.00/M + 15.00M x $15.00/M = $270.00 + $225.00
Claude Opus 4.6
Anthropic
$825.00/month
ctx 1.0M
90.0M x $5.00/M + 15.00M x $25.00/M = $450.00 + $375.00

Exchange rate: 1 USD = 6.9 CNY (as of 2026-03-21). Costs are estimates based on typical agent usage patterns. Actual costs depend on task complexity, context length, and provider billing specifics.