ภาคผนวก A — Provider, โมเดล และราคา
ภาคผนวกนี้คือ catalog ของ thClaws.cloud gateway — รายการโมเดลที่
เรียกได้เมื่อชี้ agent ไปที่ https://thclaws.cloud/gateway หรือ
รัน thClaws.cloud แบบ self-host เอง สำหรับ
desktop / CLI build จะมี catalog แยกของตัวเองที่ครอบคลุม 25 provider —
ดูที่ บทที่ 6 — Provider, โมเดล และ API key
ตารางเหล่านี้ generate มาจาก
model_catalogue.jsonซึ่งเป็น SSOT ของราคา อย่าแก้ด้วยมือ — หลังรันmake catalogueให้รันpython3 scripts/gen-appendix-a.pyแล้ว commit ไปพร้อมกัน ถ้าสองฝั่งไม่ตรงกันmake appendix-a-checkจะ fail
โครงสร้างราคา
ราคาด้านล่างคือ ราคาที่คุณจ่ายจริง — นั่นคือ upstream cost × platform markup ภายใน gateway แยกออกเป็น 2 ชั้น:
| ชั้น | สิ่งที่เก็บ | ที่อยู่ |
|---|---|---|
| อัตราใน DB row | ราคา upstream แบบดิบจาก LiteLLM (USD ต่อ token) | ตาราง model_pricing seed โดย scripts/refresh-model-catalogue.py |
| Platform markup | ตัวคูณ 1.25× ใช้ตอน meter |
env var THCLAWS_PLATFORM_MARKUP บน gateway service |
วิธีนี้หมายความว่า: ถ้า Anthropic ปรับ Opus 4.8 ขึ้นเป็น $6/M คุณแค่ อัปเดต LiteLLM (หรือรอ sync รอบหน้า) gateway จะ pick up อัตราใหม่ ในการ refresh pricing ครั้งถัดไป — markup ยังคงเป็น env var ตัวเดียว ที่ปรับแยก
ราคาทั้งหมดแสดงเป็น ดอลลาร์สหรัฐ ต่อ 1,000,000 token (1M token) ใน
DB เก็บเป็น microcent ต่อ 1k token (µ¢/kt) สูตรคือ $/M = µ¢/kt / 100,000
รายการโมเดล
ราคาที่แสดงคือราคาที่คุณจ่ายจริง — ต้นทุนต้นทางบวก markup 1.25× แล้ว
context ที่มี ? คือค่าที่ provider ไม่เคยประกาศ — เป็นค่า default ของ provider นั้นที่ใส่ไว้เพราะถ้าเว้นว่างจะทำให้ตัวเลือกโมเดลพัง ให้ถือเป็นขอบล่าง ไม่ใช่สเปกจริง
633 โมเดล จาก 14 provider
Anthropic
| Model | Context | Input ($/M) | Cached in ($/M) | Output ($/M) |
|---|---|---|---|---|
claude-3-5-haiku |
200k | $1.00 | $0.10 | $5.00 |
claude-3-5-sonnet |
200k | $3.75 | $0.38 | $18.75 |
claude-fable-5 |
1M | $12.50 | $1.25 | $62.50 |
claude-fable-5-1 |
1M | $12.50 | $1.25 | $62.50 |
claude-haiku-4-5 |
200k | $1.25 | $0.12 | $6.25 |
claude-haiku-4-5-20251001 |
200k | $1.25 | $0.12 | $6.25 |
claude-opus-4-1-20250805 |
200k | $18.75 | $1.88 | $93.75 |
claude-opus-4-20250514 |
200k | $18.75 | $1.88 | $93.75 |
claude-opus-4-5-20251101 |
200k | $6.25 | $0.62 | $31.25 |
claude-opus-4-6 |
1M | $6.25 | $0.62 | $31.25 |
claude-opus-4-7 |
1M | $6.25 | $0.62 | $31.25 |
claude-opus-4-7-1m |
1M | $6.25 | $0.62 | $31.25 |
claude-opus-4-8 |
1M | $6.25 | $0.62 | $31.25 |
claude-opus-5 |
1M | $6.25 | $0.62 | $31.25 |
claude-sonnet-4-20250514 |
200k | $3.75 | $0.38 | $18.75 |
claude-sonnet-4-5 |
200k | $3.75 | $0.38 | $18.75 |
claude-sonnet-4-5-20250929 |
200k | $3.75 | $0.38 | $18.75 |
claude-sonnet-4-6 |
1M | $3.75 | $0.38 | $18.75 |
claude-sonnet-5 |
1M | $2.50 | $0.25 | $12.50 |
OpenAI
| Model | Context | Input ($/M) | Cached in ($/M) | Output ($/M) |
|---|---|---|---|---|
gpt-3.5-turbo |
16k | $0.62 | — | $1.88 |
gpt-3.5-turbo-0125 |
16k | $0.62 | — | $1.88 |
gpt-3.5-turbo-1106 |
16k | $1.25 | — | $2.50 |
gpt-3.5-turbo-16k |
16k | $3.75 | — | $5.00 |
gpt-3.5-turbo-instruct |
4k | $1.88 | — | $2.50 |
gpt-3.5-turbo-instruct-0914 |
8k | $1.88 | — | $2.50 |
gpt-4 |
8k | $37.50 | — | $75.00 |
gpt-4-0613 |
8k | $37.50 | — | $75.00 |
gpt-4-turbo |
128k | $12.50 | — | $37.50 |
gpt-4-turbo-2024-04-09 |
128k | $12.50 | — | $37.50 |
gpt-4.1 |
1.04758M | $2.50 | $0.62 | $10.00 |
gpt-4.1-2025-04-14 |
1.04758M | $2.50 | $0.62 | $10.00 |
gpt-4.1-mini |
1M | $0.50 | $0.12 | $2.00 |
gpt-4.1-mini-2025-04-14 |
1.04758M | $0.50 | $0.12 | $2.00 |
gpt-4.1-nano |
1M | $0.12 | $0.03 | $0.50 |
gpt-4.1-nano-2025-04-14 |
1.04758M | $0.12 | $0.03 | $0.50 |
gpt-4o |
128k | $3.12 | $1.56 | $12.50 |
gpt-4o-2024-05-13 |
128k | $6.25 | — | $18.75 |
gpt-4o-2024-08-06 |
128k | $3.12 | $1.56 | $12.50 |
gpt-4o-2024-11-20 |
128k | $3.12 | $1.56 | $12.50 |
gpt-4o-mini |
128k | $0.19 | $0.09 | $0.75 |
gpt-4o-mini-2024-07-18 |
128k | $0.19 | $0.09 | $0.75 |
gpt-5 |
400k | $1.56 | $0.16 | $12.50 |
gpt-5-2025-08-07 |
272k | $1.56 | $0.16 | $12.50 |
gpt-5-mini |
400k | $0.31 | $0.03 | $2.50 |
gpt-5-mini-2025-08-07 |
272k | $0.31 | $0.03 | $2.50 |
gpt-5-nano |
400k | $0.06 | $0.0063 | $0.50 |
gpt-5-nano-2025-08-07 |
272k | $0.06 | $0.0063 | $0.50 |
gpt-5-pro |
400k | $18.75 | — | $150.00 |
gpt-5-pro-2025-10-06 |
400k | $18.75 | — | $150.00 |
gpt-5-search-api |
272k | $1.56 | $0.16 | $12.50 |
gpt-5-search-api-2025-10-14 |
272k | $1.56 | $0.16 | $12.50 |
gpt-5.1 |
400k | $1.56 | $0.16 | $12.50 |
gpt-5.1-2025-11-13 |
272k | $1.56 | $0.16 | $12.50 |
gpt-5.2 |
400k | $2.19 | $0.22 | $17.50 |
gpt-5.2-2025-12-11 |
272k | $2.19 | $0.22 | $17.50 |
gpt-5.2-pro |
400k | $26.25 | — | $210.00 |
gpt-5.2-pro-2025-12-11 |
272k | $26.25 | — | $210.00 |
gpt-5.3-codex |
400k | $2.19 | $0.22 | $17.50 |
gpt-5.4 |
1.05M | $3.12 | $0.31 | $18.75 |
gpt-5.4-2026-03-05 |
1.05M | $3.12 | $0.31 | $18.75 |
gpt-5.4-mini |
400k | $0.94 | $0.09 | $5.62 |
gpt-5.4-mini-2026-03-17 |
272k | $0.94 | $0.09 | $5.62 |
gpt-5.4-nano |
400k | $0.25 | $0.03 | $1.56 |
gpt-5.4-nano-2026-03-17 |
272k | $0.25 | $0.03 | $1.56 |
gpt-5.4-pro |
1.05M | $37.50 | $3.75 | $225.00 |
gpt-5.4-pro-2026-03-05 |
1.05M | $37.50 | $3.75 | $225.00 |
gpt-5.5 |
1.05M | $6.25 | $0.62 | $37.50 |
gpt-5.5-2026-04-23 |
1.05M | $6.25 | $0.62 | $37.50 |
gpt-5.5-pro |
1.05M | $37.50 | $3.75 | $225.00 |
gpt-5.5-pro-2026-04-23 |
1.05M | $37.50 | $3.75 | $225.00 |
gpt-5.6-luna |
1.05M | $0.25 | — | $1.50 |
gpt-5.6-sol |
1.05M | $5.00 | — | $25.00 |
gpt-5.6-terra |
1.05M | $2.50 | — | $15.00 |
gpt-6-astra |
922k | $12.50 | — | $62.50 |
o1 |
200k | $18.75 | $9.38 | $75.00 |
o1-2024-12-17 |
200k | $18.75 | $9.38 | $75.00 |
o1-mini |
128k | $1.38 | — | $5.50 |
o1-pro |
200k | $187.50 | — | $750.00 |
o1-pro-2025-03-19 |
200k | $187.50 | — | $750.00 |
o3 |
200k | $2.50 | $0.62 | $10.00 |
o3-2025-04-16 |
200k | $2.50 | $0.62 | $10.00 |
o3-deep-research |
200k | $12.50 | $3.12 | $50.00 |
o3-deep-research-2025-06-26 |
200k | $12.50 | $3.12 | $50.00 |
o3-mini |
200k | $1.38 | $0.69 | $5.50 |
o3-mini-2025-01-31 |
200k | $1.38 | $0.69 | $5.50 |
o3-pro |
200k | $25.00 | — | $100.00 |
o3-pro-2025-06-10 |
200k | $25.00 | — | $100.00 |
o4-mini |
200k | $1.38 | $0.34 | $5.50 |
o4-mini-2025-04-16 |
200k | $1.38 | $0.34 | $5.50 |
o4-mini-deep-research |
200k | $2.50 | $0.62 | $10.00 |
o4-mini-deep-research-2025-06-26 |
200k | $2.50 | $0.62 | $10.00 |
คิดเงินตามหน่วยผลลัพธ์ ไม่ใช่ตาม token
| Model | $/1k ตัวอักษร |
|---|---|
gpt-4o-mini-tts |
$0.02 |
Google (Gemini)
| Model | Context | Input ($/M) | Cached in ($/M) | Output ($/M) |
|---|---|---|---|---|
gemini-2.0-flash |
1.04858M | $0.12 | $0.03 | $0.50 |
gemini-2.0-flash-001 |
1.04858M | $0.19 | $0.05 | $0.75 |
gemini-2.0-flash-lite |
1.04858M | $0.09 | $0.02 | $0.38 |
gemini-2.0-flash-lite-001 |
1.04858M | $0.09 | $0.02 | $0.38 |
gemini-2.5-computer-use-preview-10-2025 |
128k | $1.56 | — | $12.50 |
gemini-2.5-flash |
1.04858M | $0.38 | $0.04 | $3.12 |
gemini-2.5-flash-lite |
1.04858M | $0.12 | $0.01 | $0.50 |
gemini-2.5-flash-native-audio-latest |
1.04858M | $0.62 | — | $2.50 |
gemini-2.5-flash-native-audio-preview-09-2025 |
1.04858M | $0.62 | — | $2.50 |
gemini-2.5-flash-native-audio-preview-12-2025 |
1.04858M | $0.62 | — | $2.50 |
gemini-2.5-pro |
2M | $1.56 | $0.16 | $12.50 |
gemini-3-flash-preview |
1.04858M | $0.62 | $0.06 | $3.75 |
gemini-3-pro-preview |
1.04858M | $2.50 | $0.25 | $15.00 |
gemini-3.1-flash-lite |
1.04858M | $0.31 | $0.03 | $1.88 |
gemini-3.1-flash-lite-preview |
1.04858M | $0.31 | $0.03 | $1.88 |
gemini-3.1-flash-live-preview |
131k | $0.94 | — | $5.62 |
gemini-3.1-pro-preview |
1.04858M | $2.50 | $0.25 | $15.00 |
gemini-3.1-pro-preview-customtools |
1.04858M | $2.50 | $0.25 | $15.00 |
gemini-3.5-flash |
1.04858M | $1.88 | $0.19 | $11.25 |
gemini-3.5-flash-lite |
1.04858M | $0.38 | — | $3.12 |
gemini-3.5-transcribe |
1M? | $2.50 | — | $15.00 |
gemini-3.6-flash |
1.04858M | $0.94 | — | $4.69 |
gemini-3.7-flash |
1.04858M | $0.94 | — | $4.69 |
gemini-3.8-flash |
1.04858M | $0.94 | — | $4.69 |
gemini-flash-latest |
1.04858M | $0.38 | $0.04 | $3.12 |
gemini-flash-lite-latest |
1.04858M | $0.12 | $0.01 | $0.50 |
gemini-pro-latest |
1.04858M | $1.56 | $0.16 | $12.50 |
gemini-robotics-er-1.5-preview |
1.04858M | $0.38 | free | $3.12 |
OpenRouter
| Model | Context | Input ($/M) | Cached in ($/M) | Output ($/M) |
|---|---|---|---|---|
aion-labs/aion-2.0 |
131k | $1.00 | — | $2.00 |
aion-labs/aion-3.0 |
131k | $3.75 | — | $7.50 |
aion-labs/aion-3.0-mini |
131k | $0.88 | — | $1.75 |
aion-labs/aion-rp-llama-3.1-8b |
33k | $1.00 | — | $2.00 |
amazon/nova-2-lite-v1 |
1M | $0.38 | — | $3.12 |
amazon/nova-lite-v1 |
300k | $0.07 | — | $0.30 |
amazon/nova-micro-v1 |
128k | $0.04 | — | $0.18 |
amazon/nova-premier-v1 |
1M | $3.12 | — | $15.62 |
amazon/nova-pro-v1 |
300k | $1.00 | — | $4.00 |
anthracite-org/magnum-v4-72b |
16k | $3.12 | — | $6.25 |
anthropic/claude-3-haiku |
200k | $0.31 | — | $1.56 |
anthropic/claude-fable-5 |
1M | $12.50 | $1.25 | $62.50 |
anthropic/claude-fable-5.1 |
1M | $12.50 | — | $62.50 |
anthropic/claude-haiku-4.5 |
200k | $1.25 | $0.12 | $6.25 |
anthropic/claude-opus-4 |
200k | $18.75 | $1.88 | $93.75 |
anthropic/claude-opus-4.1 |
200k | $18.75 | $1.88 | $93.75 |
anthropic/claude-opus-4.5 |
200k | $6.25 | $0.62 | $31.25 |
anthropic/claude-opus-4.6 |
1M | $6.25 | $0.62 | $31.25 |
anthropic/claude-opus-4.7 |
1M | $6.25 | $0.62 | $31.25 |
anthropic/claude-opus-4.8 |
1M | $6.25 | — | $31.25 |
anthropic/claude-opus-5 |
1M | $6.25 | — | $31.25 |
anthropic/claude-sonnet-4 |
1M | $3.75 | $0.38 | $18.75 |
anthropic/claude-sonnet-4.5 |
1M | $3.75 | $0.38 | $18.75 |
anthropic/claude-sonnet-4.6 |
1M | $3.75 | $0.38 | $18.75 |
anthropic/claude-sonnet-5 |
1M | $2.50 | — | $12.50 |
arcee-ai/trinity-large-thinking |
262k | $0.31 | — | $1.00 |
baidu/ernie-4.5-vl-424b-a47b |
123k | $0.53 | — | $1.56 |
bytedance-seed/seed-1.6 |
262k | $0.31 | — | $2.50 |
bytedance-seed/seed-1.6-flash |
262k | $0.09 | — | $0.38 |
bytedance-seed/seed-2-1-turbo |
262k | $0.62 | — | $3.12 |
bytedance-seed/seed-2.0-code |
262k | $0.62 | — | $3.75 |
bytedance-seed/seed-2.0-lite |
262k | $0.31 | — | $2.50 |
bytedance-seed/seed-2.0-mini |
262k | $0.12 | — | $0.50 |
bytedance/ui-tars-1.5-7b |
128k | $0.12 | — | $0.25 |
cognitivecomputations/dolphin-mistral-24b-venice-edition |
128k | $0.25 | — | $1.12 |
cohere/command-a |
256k | $3.12 | — | $12.50 |
cohere/command-r-08-2024 |
128k | $0.19 | — | $0.75 |
cohere/command-r-plus-08-2024 |
128k | $3.12 | — | $12.50 |
cohere/command-r7b-12-2024 |
128k | $0.05 | — | $0.19 |
cohere/north-mini-code:free |
256k | free | — | free |
deepseek/deepseek-chat |
164k | $0.40 | — | $1.11 |
deepseek/deepseek-chat-v3-0324 |
164k | $0.31 | — | $1.25 |
deepseek/deepseek-chat-v3.1 |
33k | $0.69 | — | $2.06 |
deepseek/deepseek-r1 |
64k | $0.88 | — | $3.12 |
deepseek/deepseek-r1-0528 |
164k | $0.62 | — | $2.69 |
deepseek/deepseek-r1-distill-llama-70b |
131k | $1.00 | — | $1.00 |
deepseek/deepseek-v3.1-terminus |
164k | $0.34 | — | $1.25 |
deepseek/deepseek-v3.2 |
131k | $0.34 | — | $0.50 |
deepseek/deepseek-v3.2-exp |
164k | $0.34 | — | $0.51 |
deepseek/deepseek-v4-flash |
1.04858M | $0.11 | — | $0.21 |
deepseek/deepseek-v4-flash-0731 |
1.04858M | $0.08 | — | $0.22 |
deepseek/deepseek-v4-flash-vision-exp |
1.04858M | $0.28 | — | $0.83 |
deepseek/deepseek-v4-pro |
1.04858M | $1.06 | — | $2.12 |
deepseek/deepseek-v4-pro-0813 |
1.04858M | $0.72 | — | $2.17 |
dots-studio/dots-3-note-preview:free |
512k | free | — | free |
google/gemini-2.5-flash |
1.04858M | $0.38 | — | $3.12 |
google/gemini-2.5-flash-lite |
1.04858M | $0.12 | $0.01 | $0.50 |
google/gemini-2.5-pro |
1.04858M | $1.56 | — | $12.50 |
google/gemini-2.5-pro-preview |
1.04858M | $1.56 | — | $12.50 |
google/gemini-2.5-pro-preview-05-06 |
1.04858M | $1.56 | — | $12.50 |
google/gemini-3-flash-preview |
1.04858M | $0.62 | $0.06 | $3.75 |
google/gemini-3-pro-image |
131k | $2.50 | — | $15.00 |
google/gemini-3.1-flash-image |
131k | $0.62 | — | $3.75 |
google/gemini-3.1-flash-lite |
1.04858M | $0.31 | $0.03 | $1.88 |
google/gemini-3.1-flash-lite-image |
66k | $0.31 | — | $1.88 |
google/gemini-3.1-flash-lite-preview |
1.04858M | $0.31 | $0.03 | $1.88 |
google/gemini-3.1-pro-preview |
1.04858M | $2.50 | $0.25 | $15.00 |
google/gemini-3.1-pro-preview-customtools |
1.04858M | $2.50 | $0.25 | $15.00 |
google/gemini-3.5-flash |
1.04858M | $1.88 | $0.19 | $11.25 |
google/gemini-3.5-flash-lite |
1.04858M | $0.38 | — | $3.12 |
google/gemini-3.6-flash |
1.04858M | $0.94 | — | $4.69 |
google/gemini-3.7-flash |
1.04858M | $0.94 | — | $4.69 |
google/gemini-3.8-flash |
1.04858M | $0.94 | — | $4.69 |
google/gemma-2-27b-it |
8k | $0.81 | — | $0.81 |
google/gemma-3-12b-it |
131k | $0.06 | — | $0.19 |
google/gemma-3-27b-it |
131k | $0.10 | — | $0.56 |
google/gemma-3-4b-it |
131k | $0.06 | — | $0.12 |
google/gemma-4-26b-a4b-it |
262k | $0.09 | — | $0.43 |
google/gemma-4-26b-a4b-it:free |
262k | free | — | free |
google/gemma-4-31b-it |
262k | $0.11 | — | $0.43 |
google/gemma-4-31b-it:free |
262k | free | — | free |
gryphe/mythomax-l2-13b |
4k | $0.07 | — | $0.07 |
ibm-granite/granite-4.0-h-micro |
131k | $0.02 | — | $0.14 |
ibm-granite/granite-4.2-8b |
131k | $0.12 | — | $0.19 |
inception/mercury-2 |
128k | $0.31 | $0.03 | $0.94 |
inception/mercury-2.5-preview |
260k | $0.05 | — | $0.19 |
inclusionai/ling-3.0-flash |
262k | $0.03 | — | $0.08 |
inclusionai/ling-3.0-flash-fin |
262k | $0.07 | — | $0.22 |
inclusionai/ling-3.0-flash-fin:free |
262k | free | — | free |
inclusionai/ling-3.0-flash-sante:free |
262k | free | — | free |
kwaipilot/kat-coder-pro-v2 |
256k | $0.38 | — | $1.50 |
kwaipilot/kat-coder-pro-v2.5 |
256k | $0.93 | — | $3.70 |
liquid/lfm-2.5-2.6b:free |
66k | free | — | free |
mancer/weaver |
8k | $0.50 | — | $0.94 |
meituan/longcat-2.0 |
1.04876M | $0.38 | — | $1.50 |
meta-llama/llama-3.1-70b-instruct |
131k | $0.50 | — | $0.50 |
meta-llama/llama-3.1-8b-instruct |
16k | $0.06 | — | $0.10 |
meta-llama/llama-3.2-1b-instruct |
60k | $0.03 | — | $0.25 |
meta-llama/llama-3.2-3b-instruct |
80k | $0.06 | — | $0.41 |
meta-llama/llama-3.3-70b-instruct |
131k | $0.12 | — | $0.40 |
meta-llama/llama-4-maverick |
1.04858M | $0.25 | — | $0.87 |
meta-llama/llama-4-scout |
328k | $0.12 | — | $0.38 |
meta-llama/llama-guard-4-12b |
164k | $0.22 | — | $0.22 |
meta/muse-glimmer-30b |
131k | $0.38 | — | $1.38 |
meta/muse-spark-1.1 |
1.04858M | $1.56 | — | $5.31 |
meta/muse-spark-1.2 |
1.04858M | $1.56 | — | $5.31 |
meta/muse-spark-1.2-contributor |
1.04858M | $0.12 | — | $0.25 |
meta/muse-spark-1.3 |
1.04858M | $1.56 | — | $5.31 |
meta/muse-spark-1.3-contributor |
1.04858M | $0.12 | — | $0.25 |
microsoft/phi-4 |
16k | $0.09 | — | $0.18 |
microsoft/wizardlm-2-8x22b |
66k | $0.78 | — | $0.78 |
minimax/minimax-01 |
1.00019M | $0.25 | — | $1.38 |
minimax/minimax-m1 |
1M | $0.69 | — | $2.75 |
minimax/minimax-m2 |
197k | $0.32 | — | $1.27 |
minimax/minimax-m2-her |
66k | $0.38 | — | $1.50 |
minimax/minimax-m2.1 |
197k | $0.38 | free | $1.50 |
minimax/minimax-m2.5 |
197k | $0.34 | $0.19 | $1.35 |
minimax/minimax-m2.7 |
197k | $0.38 | — | $1.50 |
minimax/minimax-m2.7:free |
197k | free | — | free |
minimax/minimax-m3 |
1.04858M | $0.38 | — | $1.50 |
minimax/minimax-m3:free |
1.04858M | free | — | free |
mistralai/codestral-2508 |
256k | $0.38 | — | $1.12 |
mistralai/devstral-2512 |
262k | $0.50 | — | $2.50 |
mistralai/ministral-14b-2512 |
262k | $0.25 | — | $0.25 |
mistralai/ministral-3b-2512 |
131k | $0.12 | — | $0.12 |
mistralai/ministral-8b-2512 |
262k | $0.19 | — | $0.19 |
mistralai/mistral-large |
128k | $2.50 | — | $7.50 |
mistralai/mistral-large-2407 |
131k | $2.50 | — | $7.50 |
mistralai/mistral-large-2512 |
262k | $0.62 | — | $1.88 |
mistralai/mistral-medium-3 |
131k | $0.50 | — | $2.50 |
mistralai/mistral-medium-3-5 |
262k | $1.88 | — | $9.38 |
mistralai/mistral-medium-3.1 |
131k | $0.50 | — | $2.50 |
mistralai/mistral-nemo |
131k | $0.02 | — | $0.04 |
mistralai/mistral-saba |
33k | $0.25 | — | $0.75 |
mistralai/mistral-small-24b-instruct-2501 |
33k | $0.06 | — | $0.10 |
mistralai/mistral-small-2603 |
262k | $0.19 | — | $0.75 |
mistralai/mistral-small-3.1-24b-instruct |
128k | $0.44 | — | $0.69 |
mistralai/mistral-small-3.2-24b-instruct |
256k | $0.09 | — | $0.25 |
mistralai/mixtral-8x22b-instruct |
66k | $2.50 | — | $7.50 |
mistralai/voxtral-small-24b-2507 |
32k | $0.12 | — | $0.38 |
moonshotai/kimi-k2 |
131k | $0.71 | — | $2.88 |
moonshotai/kimi-k2-0905 |
262k | $0.75 | — | $3.12 |
moonshotai/kimi-k2-thinking |
262k | $0.75 | — | $3.12 |
moonshotai/kimi-k2.5 |
262k | $0.56 | $0.12 | $2.81 |
moonshotai/kimi-k2.6 |
256k | $1.19 | — | $5.00 |
moonshotai/kimi-k2.7-code |
262k | $0.83 | — | $4.25 |
moonshotai/kimi-k3 |
1.04858M | $3.75 | — | $18.75 |
morph/morph-v3-fast |
82k | $1.00 | — | $1.50 |
morph/morph-v3-large |
262k | $1.12 | — | $2.38 |
nex-agi/nex-n2-mini |
262k | $0.03 | — | $0.12 |
nex-agi/nex-n2-pro |
262k | $0.31 | — | $1.25 |
nousresearch/hermes-3-llama-3.1-405b |
131k | $1.25 | — | $1.25 |
nousresearch/hermes-3-llama-3.1-70b |
131k | $0.88 | — | $0.88 |
nousresearch/hermes-4-405b |
131k | $1.25 | — | $3.75 |
nousresearch/hermes-4-70b |
131k | $0.16 | — | $0.50 |
nvidia/nemotron-3-nano-30b-a3b |
262k | $0.06 | — | $0.25 |
nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free |
256k | free | — | free |
nvidia/nemotron-3-super-120b-a12b |
262k | $0.11 | — | $0.50 |
nvidia/nemotron-3-super-120b-a12b:free |
262k | free | — | free |
nvidia/nemotron-3-ultra-550b-a55b |
1M | $0.78 | — | $3.91 |
nvidia/nemotron-3-ultra-550b-a55b:free |
1M | free | — | free |
nvidia/nemotron-3.5-content-safety |
131k | $0.25 | — | $0.25 |
nvidia/nemotron-3.5-content-safety:free |
128k | free | — | free |
nvidia/nemotron-3.5-lightning |
262k | $0.10 | — | $0.25 |
nvidia/nemotron-3.5-lightning:free |
1M | free | — | free |
openai/gpt-3.5-turbo |
16k | $0.62 | — | $1.88 |
openai/gpt-3.5-turbo-0613 |
4k | $1.25 | — | $2.50 |
openai/gpt-3.5-turbo-16k |
16k | $3.75 | — | $5.00 |
openai/gpt-3.5-turbo-instruct |
4k | $1.88 | — | $2.50 |
openai/gpt-4 |
8k | $37.50 | — | $75.00 |
openai/gpt-4-turbo |
128k | $12.50 | — | $37.50 |
openai/gpt-4-turbo-preview |
128k | $12.50 | — | $37.50 |
openai/gpt-4.1 |
1.04758M | $2.50 | $0.62 | $10.00 |
openai/gpt-4.1-mini |
1.04758M | $0.50 | $0.12 | $2.00 |
openai/gpt-4.1-nano |
1.04758M | $0.12 | $0.03 | $0.50 |
openai/gpt-4o |
128k | $3.12 | — | $12.50 |
openai/gpt-4o-2024-05-13 |
128k | $6.25 | — | $18.75 |
openai/gpt-4o-2024-08-06 |
128k | $3.12 | $1.56 | $12.50 |
openai/gpt-4o-2024-11-20 |
128k | $3.12 | $1.56 | $12.50 |
openai/gpt-4o-mini |
128k | $0.19 | $0.09 | $0.75 |
openai/gpt-4o-mini-2024-07-18 |
128k | $0.19 | $0.09 | $0.75 |
openai/gpt-5 |
400k | $1.56 | $0.16 | $12.50 |
openai/gpt-5-mini |
400k | $0.31 | $0.03 | $2.50 |
openai/gpt-5-nano |
400k | $0.06 | $0.0063 | $0.50 |
openai/gpt-5-pro |
400k | $18.75 | — | $150.00 |
openai/gpt-5.1 |
400k | $1.56 | $0.16 | $12.50 |
openai/gpt-5.1-codex |
400k | $1.56 | $0.16 | $12.50 |
openai/gpt-5.1-codex-max |
400k | $1.56 | $0.16 | $12.50 |
openai/gpt-5.1-codex-mini |
400k | $0.31 | $0.03 | $2.50 |
openai/gpt-5.2 |
400k | $2.19 | $0.22 | $17.50 |
openai/gpt-5.2-chat |
128k | $2.19 | $0.22 | $17.50 |
openai/gpt-5.2-codex |
400k | $2.19 | $0.22 | $17.50 |
openai/gpt-5.2-pro |
400k | $26.25 | — | $210.00 |
openai/gpt-5.3-codex |
400k | $2.19 | $0.22 | $17.50 |
openai/gpt-5.4 |
1.05M | $3.12 | $0.31 | $18.75 |
openai/gpt-5.4-mini |
400k | $0.94 | $0.09 | $5.62 |
openai/gpt-5.4-nano |
400k | $0.25 | $0.03 | $1.56 |
openai/gpt-5.4-pro |
1.05M | $37.50 | $3.75 | $225.00 |
openai/gpt-5.5 |
1.05M | $6.25 | $0.62 | $37.50 |
openai/gpt-5.5-pro |
1.05M | $37.50 | $3.75 | $225.00 |
openai/gpt-5.6-luna |
1.05M | $0.25 | — | $1.50 |
openai/gpt-5.6-luna-pro |
1.05M | $0.25 | — | $1.50 |
openai/gpt-5.6-sol |
1.05M | $2.50 | — | $12.50 |
openai/gpt-5.6-sol-pro |
1.05M | $2.50 | — | $12.50 |
openai/gpt-5.6-terra |
1.05M | $2.50 | — | $15.00 |
openai/gpt-5.6-terra-pro |
1.05M | $2.50 | — | $15.00 |
openai/gpt-6-astra |
1.05M | $12.50 | — | $62.50 |
openai/gpt-6-astra-pro |
1.05M | $12.50 | — | $62.50 |
openai/gpt-chat-latest |
400k | $6.25 | — | $37.50 |
openai/gpt-oss-120b |
131k | $0.05 | — | $0.21 |
openai/gpt-oss-20b |
131k | $0.04 | — | $0.16 |
openai/gpt-oss-safeguard-20b |
131k | $0.09 | — | $0.38 |
openai/o1 |
200k | $18.75 | $9.38 | $75.00 |
openai/o1-pro |
200k | $187.50 | — | $750.00 |
openai/o3 |
200k | $2.50 | $0.62 | $10.00 |
openai/o3-mini |
200k | $1.38 | — | $5.50 |
openai/o3-mini-high |
200k | $1.38 | — | $5.50 |
openai/o3-pro |
200k | $25.00 | — | $100.00 |
openai/o4-mini |
200k | $1.38 | $0.34 | $5.50 |
openai/o4-mini-high |
200k | $1.38 | — | $5.50 |
openrouter/auto-beta |
2M | $-1250000 | — | $-1250000 |
openrouter/bodybuilder |
128k | $-1250000 | — | $-1250000 |
openrouter/free |
200k | free | — | free |
openrouter/fusion |
1M | $-1250000 | — | $-1250000 |
openrouter/pareto-code |
200k | $-1250000 | — | $-1250000 |
perceptron/perceptron-mk1 |
33k | $0.19 | — | $1.88 |
perplexity/sonar |
127k | $1.25 | — | $1.25 |
perplexity/sonar-deep-research |
128k | $2.50 | — | $10.00 |
perplexity/sonar-pro |
200k | $3.75 | — | $18.75 |
perplexity/sonar-pro-search |
200k | $3.75 | — | $18.75 |
perplexity/sonar-reasoning-pro |
128k | $2.50 | — | $10.00 |
poolside/laguna-s-2.1 |
1.04858M | $0.11 | — | $0.22 |
poolside/laguna-s-2.1:free |
262k | free | — | free |
poolside/laguna-xs-2.1 |
262k | $0.07 | — | $0.15 |
poolside/laguna-xs-2.1:free |
262k | free | — | free |
qwen/qwen-2.5-72b-instruct |
33k | $0.45 | — | $0.50 |
qwen/qwen-2.5-7b-instruct |
33k | $0.12 | — | $0.25 |
qwen/qwen-2.5-coder-32b-instruct |
33k | $0.83 | — | $1.25 |
qwen/qwen-plus |
1M | $0.33 | — | $0.98 |
qwen/qwen-plus-2025-07-28 |
1M | $0.33 | — | $0.98 |
qwen/qwen2.5-vl-72b-instruct |
32k | $1.00 | — | $1.25 |
qwen/qwen3-14b |
41k | $0.15 | — | $0.30 |
qwen/qwen3-235b-a22b |
131k | $0.57 | — | $2.27 |
qwen/qwen3-235b-a22b-2507 |
262k | $0.11 | — | $0.69 |
qwen/qwen3-235b-a22b-thinking-2507 |
131k | $0.29 | — | $2.88 |
qwen/qwen3-30b-a3b |
41k | $0.15 | — | $0.62 |
qwen/qwen3-30b-a3b-instruct-2507 |
262k | $0.06 | — | $0.24 |
qwen/qwen3-30b-a3b-thinking-2507 |
131k | $0.25 | — | $3.00 |
qwen/qwen3-32b |
41k | $0.10 | — | $0.35 |
qwen/qwen3-8b |
41k | $0.15 | — | $0.57 |
qwen/qwen3-coder |
262k | $0.38 | — | $1.25 |
qwen/qwen3-coder-30b-a3b-instruct |
160k | $0.09 | — | $0.35 |
qwen/qwen3-coder-flash |
1M | $0.24 | — | $1.22 |
qwen/qwen3-coder-next |
262k | $0.15 | — | $1.00 |
qwen/qwen3-coder-plus |
1M | $0.81 | — | $4.06 |
qwen/qwen3-max |
262k | $0.98 | — | $4.88 |
qwen/qwen3-max-thinking |
262k | $0.98 | — | $4.88 |
qwen/qwen3-next-80b-a3b-instruct |
262k | $0.12 | — | $1.38 |
qwen/qwen3-next-80b-a3b-thinking |
131k | $0.19 | — | $1.50 |
qwen/qwen3-vl-235b-a22b-instruct |
262k | $0.26 | — | $2.38 |
qwen/qwen3-vl-235b-a22b-thinking |
131k | $0.50 | — | $5.00 |
qwen/qwen3-vl-30b-a3b-instruct |
131k | $0.19 | — | $0.75 |
qwen/qwen3-vl-30b-a3b-thinking |
131k | $0.25 | — | $3.00 |
qwen/qwen3-vl-32b-instruct |
131k | $0.13 | — | $0.52 |
qwen/qwen3-vl-8b-instruct |
131k | $0.15 | — | $0.57 |
qwen/qwen3-vl-8b-thinking |
131k | $0.22 | — | $2.62 |
qwen/qwen3.5-122b-a10b |
262k | $0.36 | — | $3.00 |
qwen/qwen3.5-27b |
262k | $0.24 | — | $1.95 |
qwen/qwen3.5-35b-a3b |
262k | $0.10 | — | $0.94 |
qwen/qwen3.5-397b-a17b |
262k | $0.69 | — | $4.38 |
qwen/qwen3.5-9b |
262k | $0.12 | — | $0.19 |
qwen/qwen3.5-flash-02-23 |
1M | $0.08 | — | $0.33 |
qwen/qwen3.5-plus-02-15 |
1M | $0.33 | — | $1.95 |
qwen/qwen3.5-plus-20260420 |
1M | $0.38 | — | $2.25 |
qwen/qwen3.6-27b |
262k | $0.38 | — | $2.50 |
qwen/qwen3.6-35b-a3b |
262k | $0.12 | — | $1.12 |
qwen/qwen3.6-flash |
1M | $0.23 | — | $1.41 |
qwen/qwen3.6-max-preview |
262k | $1.28 | — | $7.70 |
qwen/qwen3.6-plus |
1M | $0.41 | — | $2.44 |
qwen/qwen3.7-flash |
1M | $0.04 | — | $0.16 |
qwen/qwen3.7-max |
1M | $1.84 | — | $5.53 |
qwen/qwen3.7-plus |
1M | $0.40 | — | $1.60 |
qwen/qwen3.8-2.4t-a95b |
1.04858M | $2.50 | — | $7.50 |
qwen/qwen3.8-27b |
1M | $0.53 | — | $3.75 |
qwen/qwen3.8-flash |
1M | $0.19 | — | $0.59 |
qwen/qwen3.8-max-0902 |
1M | $2.50 | — | $7.50 |
rekaai/reka-edge |
16k | $0.12 | — | $0.12 |
rekaai/reka-flash-3 |
66k | $0.12 | — | $0.25 |
relace/relace-apply-3 |
256k | $1.06 | — | $1.56 |
relace/relace-search |
256k | $1.25 | — | $3.75 |
sakana/fugu-ultra |
1M | $6.25 | — | $37.50 |
sakana/sakana-namazu |
262k | $1.19 | — | $5.00 |
sao10k/l3-lunaris-8b |
8k | $0.05 | — | $0.06 |
sao10k/l3.1-euryale-70b |
131k | $1.06 | — | $1.06 |
sao10k/l3.3-euryale-70b |
131k | $0.81 | — | $0.94 |
stepfun/step-3.5-flash |
262k | $0.12 | — | $0.38 |
stepfun/step-3.7-flash |
256k | $0.25 | — | $1.44 |
tencent/hunyuan-a13b-instruct |
131k | $0.18 | — | $0.71 |
tencent/hy-mt2-1.8b |
8k | $0.05 | — | $0.22 |
tencent/hy-mt2-30b-a3b |
8k | $0.09 | — | $0.37 |
tencent/hy-mt2-7b |
8k | $0.09 | — | $0.37 |
tencent/hy3 |
262k | $0.10 | — | $0.41 |
tencent/hy3-preview |
262k | $0.22 | — | $0.75 |
tencent/hy4-preview |
1.04858M | $1.04 | — | $3.13 |
thedrummer/cydonia-24b-v4.1 |
131k | $0.38 | — | $0.62 |
thedrummer/skyfall-36b-v2 |
33k | $0.69 | — | $1.00 |
thedrummer/unslopnemo-12b |
33k | $0.50 | — | $0.50 |
thinkingmachines/inkling |
1.04858M | $1.25 | — | $5.06 |
thinkingmachines/inkling-small |
524k | $0.56 | — | $1.50 |
thinkingmachines/inkling-small:free |
262k | free | — | free |
thinkingmachines/inkling:free |
262k | free | — | free |
undi95/remm-slerp-l2-13b |
6k | $0.44 | — | $0.81 |
upstage/solar-pro-3 |
131k | $0.19 | — | $0.75 |
upstage/solar-pro4 |
524k | $0.04 | — | $0.15 |
writer/palmyra-x5 |
1.04M | $0.75 | — | $7.50 |
x-ai/grok-4.20 |
2M | $1.56 | — | $3.12 |
x-ai/grok-4.20-multi-agent |
2M | $1.56 | — | $3.12 |
x-ai/grok-4.3 |
1M | $1.56 | — | $3.12 |
x-ai/grok-4.5 |
500k | $2.50 | — | $7.50 |
x-ai/grok-4.6 |
500k | $2.50 | — | $7.50 |
x-ai/grok-build-0.1 |
256k | $1.25 | — | $2.50 |
xiaomi/mimo-v2.5 |
1.04858M | $0.18 | $0.10 | $0.35 |
xiaomi/mimo-v2.5-pro |
1.04858M | $0.54 | $0.25 | $1.09 |
z-ai/glm-4.5 |
131k | $0.75 | — | $2.75 |
z-ai/glm-4.5-air |
131k | $0.16 | — | $1.06 |
z-ai/glm-4.5v |
66k | $0.75 | — | $2.25 |
z-ai/glm-4.6 |
205k | $0.69 | — | $2.75 |
z-ai/glm-4.6v |
131k | $0.38 | — | $1.12 |
z-ai/glm-4.7 |
203k | $0.50 | free | $2.19 |
z-ai/glm-4.7-flash |
203k | $0.07 | free | $0.50 |
z-ai/glm-5 |
203k | $0.75 | — | $2.40 |
z-ai/glm-5-turbo |
203k | $1.50 | — | $5.00 |
z-ai/glm-5.1 |
203k | $1.21 | — | $3.79 |
z-ai/glm-5.2 |
1.04858M | $1.21 | — | $3.79 |
z-ai/glm-5.2:free |
256k | free | — | free |
z-ai/glm-5.3 |
1.04858M | $1.75 | — | $5.50 |
z-ai/glm-5.3-flash |
1.31072M | $0.09 | — | $0.31 |
z-ai/glm-5v-turbo |
203k | $1.50 | — | $5.00 |
~anthropic/claude-fable-latest |
1M | $12.50 | — | $62.50 |
~anthropic/claude-haiku-latest |
200k | $1.25 | — | $6.25 |
~anthropic/claude-opus-latest |
1M | $6.25 | — | $31.25 |
~anthropic/claude-sonnet-latest |
1M | $2.50 | — | $12.50 |
~google/gemini-flash-latest |
1.04858M | $0.94 | $0.04 | $4.69 |
~google/gemini-pro-latest |
1.04858M | $2.50 | $0.16 | $15.00 |
~moonshotai/kimi-latest |
256k | $3.19 | — | $15.94 |
~openai/gpt-latest |
1.05M | $2.50 | — | $12.50 |
~openai/gpt-mini-latest |
400k | $0.94 | — | $5.62 |
DashScope (Alibaba)
| Model | Context | Input ($/M) | Output ($/M) |
|---|---|---|---|
dashscope/ZHIPU/GLM-5.3 |
262k? | $1.75 | $5.50 |
dashscope/deepseek-v3.2 |
164k | $0.36 | $0.54 |
dashscope/deepseek-v4-flash |
1M | $0.25 | $0.50 |
dashscope/deepseek-v4-flash-0731 |
1.04858M | $0.25 | $0.50 |
dashscope/deepseek-v4-pro |
1M | $3.00 | $6.00 |
dashscope/deepseek-v4-pro-0813 |
1M | $3.00 | $6.00 |
dashscope/glm-5.1 |
205k | $1.07 | $3.94 |
dashscope/glm-5.2 |
1.04858M | $1.07 | $3.94 |
dashscope/glm-5.2-fast-preview |
131k | $1.07 | $3.94 |
dashscope/kimi-k2.7-code |
262k | $0.72 | $3.76 |
dashscope/kimi-k3 |
262k? | $3.75 | $18.75 |
dashscope/kimi/kimi-k3 |
262k? | $3.75 | $18.75 |
dashscope/qvq-max |
131k | $1.50 | $6.00 |
dashscope/qwen-coder-plus |
262k? | $7.50 | $75.00 |
dashscope/qwen-flash |
998k | $0.31 | $2.50 |
dashscope/qwen-flash-character |
8k | $0.31 | $2.50 |
dashscope/qwen-max |
31k | $2.00 | $8.00 |
dashscope/qwen-mt-flash |
131k | $0.20 | $0.61 |
dashscope/qwen-mt-lite |
131k | $0.15 | $0.45 |
dashscope/qwen-mt-plus |
131k | $3.08 | $9.21 |
dashscope/qwen-mt-turbo |
131k | $0.20 | $0.61 |
dashscope/qwen-omni-turbo |
33k | $0.09 | $0.34 |
dashscope/qwen-plus |
129k | $0.50 | $1.50 |
dashscope/qwen-plus-2025-01-25 |
129k | $0.50 | $1.50 |
dashscope/qwen-plus-2025-04-28 |
129k | $0.50 | $1.50 |
dashscope/qwen-plus-2025-07-14 |
129k | $0.50 | $1.50 |
dashscope/qwen-plus-2025-09-11 |
998k | $1.50 | $4.50 |
dashscope/qwen-plus-2025-12-01 |
1M | $1.50 | $4.50 |
dashscope/qwen-plus-character |
33k | $1.50 | $4.50 |
dashscope/qwen-plus-latest |
998k | $1.50 | $4.50 |
dashscope/qwen-turbo |
129k | $0.06 | $0.25 |
dashscope/qwen-vl-max |
131k | $1.00 | $4.00 |
dashscope/qwen-vl-ocr-2025-11-20 |
31k | $0.09 | $0.20 |
dashscope/qwen-vl-plus |
8k | $0.26 | $0.79 |
dashscope/qwen3-14b |
131k | $0.44 | $1.75 |
dashscope/qwen3-235b-a22b |
131k | $0.88 | $3.50 |
dashscope/qwen3-235b-a22b-instruct-2507 |
129k | $0.88 | $3.50 |
dashscope/qwen3-235b-a22b-thinking-2507 |
262k | $0.88 | $10.50 |
dashscope/qwen3-30b-a3b |
129k | $0.25 | $1.00 |
dashscope/qwen3-30b-a3b-instruct-2507 |
262k | $0.25 | $1.00 |
dashscope/qwen3-30b-a3b-thinking-2507 |
82k | $0.25 | $1.00 |
dashscope/qwen3-32b |
131k | $0.20 | $0.80 |
dashscope/qwen3-8b |
131k | $0.22 | $0.88 |
dashscope/qwen3-coder-480b-a35b-instruct |
262k | $7.50 | $75.00 |
dashscope/qwen3-coder-flash |
998k | $2.00 | $12.00 |
dashscope/qwen3-coder-next |
262k | $7.50 | $75.00 |
dashscope/qwen3-coder-plus |
998k | $7.50 | $75.00 |
dashscope/qwen3-coder-plus-2025-07-22 |
998k | $7.50 | $75.00 |
dashscope/qwen3-coder-plus-2025-09-23 |
1M | $7.50 | $75.00 |
dashscope/qwen3-livetranslate-flash-realtime |
53k | $12.50 | $47.50 |
dashscope/qwen3-livetranslate-flash-realtime-2025-09-22 |
49k | $12.50 | $47.50 |
dashscope/qwen3-max |
258k | $3.75 | $18.75 |
dashscope/qwen3-max-2025-09-23 |
256k | $3.75 | $18.75 |
dashscope/qwen3-max-2026-01-23 |
258k | $3.75 | $18.75 |
dashscope/qwen3-max-preview |
258k | $3.75 | $18.75 |
dashscope/qwen3-next-80b-a3b-instruct |
262k | $0.19 | $1.50 |
dashscope/qwen3-next-80b-a3b-thinking |
262k | $0.19 | $1.50 |
dashscope/qwen3-omni-30b-a3b-captioner |
31k | $4.76 | $3.83 |
dashscope/qwen3-omni-flash |
66k | $0.54 | $2.07 |
dashscope/qwen3-omni-flash-2025-09-15 |
262k? | $0.54 | $2.07 |
dashscope/qwen3-omni-flash-2025-12-01 |
16k | $0.54 | $2.07 |
dashscope/qwen3-omni-flash-realtime |
66k | $0.65 | $2.49 |
dashscope/qwen3-omni-flash-realtime-2025-09-15 |
57k | $0.65 | $2.49 |
dashscope/qwen3-omni-flash-realtime-2025-12-01 |
57k | $0.65 | $2.49 |
dashscope/qwen3-vl-235b-a22b-instruct |
131k | $0.50 | $2.00 |
dashscope/qwen3-vl-235b-a22b-thinking |
131k | $0.50 | $5.00 |
dashscope/qwen3-vl-flash |
256k | $0.31 | $2.50 |
dashscope/qwen3-vl-flash-2025-10-15 |
256k | $0.31 | $2.50 |
dashscope/qwen3-vl-flash-2026-01-22 |
256k | $0.31 | $2.50 |
dashscope/qwen3-vl-plus |
260k | $0.25 | $2.00 |
dashscope/qwen3-vl-plus-2025-09-23 |
256k | $0.25 | $2.00 |
dashscope/qwen3-vl-plus-2025-12-19 |
256k | $0.25 | $2.00 |
dashscope/qwen3.5-122b-a10b |
262k | $0.50 | $4.00 |
dashscope/qwen3.5-27b |
262k | $0.38 | $3.00 |
dashscope/qwen3.5-35b-a3b |
262k | $0.31 | $2.50 |
dashscope/qwen3.5-397b-a17b |
262k | $0.75 | $4.50 |
dashscope/qwen3.5-flash |
1M | $0.12 | $0.50 |
dashscope/qwen3.5-flash-2026-02-23 |
1M | $0.12 | $0.50 |
dashscope/qwen3.5-livetranslate-flash-realtime |
131k | $9.38 | $37.50 |
dashscope/qwen3.5-livetranslate-flash-realtime-2026-05-19 |
131k | $9.38 | $37.50 |
dashscope/qwen3.5-omni-flash |
197k | $0.50 | $2.75 |
dashscope/qwen3.5-omni-flash-2026-03-15 |
197k | $0.50 | $2.75 |
dashscope/qwen3.5-omni-flash-realtime |
197k | $0.69 | $4.12 |
dashscope/qwen3.5-omni-flash-realtime-2026-03-15 |
197k | $0.69 | $4.12 |
dashscope/qwen3.5-omni-plus |
197k | $1.75 | $10.38 |
dashscope/qwen3.5-omni-plus-2026-03-15 |
197k | $1.75 | $10.38 |
dashscope/qwen3.5-omni-plus-realtime |
197k | $2.62 | $15.50 |
dashscope/qwen3.5-omni-plus-realtime-2026-03-15 |
197k | $2.62 | $15.50 |
dashscope/qwen3.5-plus |
992k | $0.62 | $3.75 |
dashscope/qwen3.5-plus-2026-02-15 |
1M | $0.62 | $3.75 |
dashscope/qwen3.5-plus-2026-04-20 |
1M | $0.62 | $3.75 |
dashscope/qwen3.6-27b |
262k | $0.75 | $4.50 |
dashscope/qwen3.6-35b-a3b |
262k | $0.47 | $2.81 |
dashscope/qwen3.6-flash |
1M | $1.25 | $5.00 |
dashscope/qwen3.6-flash-2026-04-16 |
1M | $1.25 | $5.00 |
dashscope/qwen3.6-max-preview |
262k | $2.50 | $15.00 |
dashscope/qwen3.6-plus |
1M | $2.50 | $7.50 |
dashscope/qwen3.6-plus-2026-04-02 |
1M | $2.50 | $7.50 |
dashscope/qwen3.7-flash |
1M | $0.25 | $1.00 |
dashscope/qwen3.7-flash-2026-07-15 |
131k | $0.25 | $1.00 |
dashscope/qwen3.7-max |
992k | $3.12 | $9.38 |
dashscope/qwen3.7-max-2026-05-17 |
131k | $3.12 | $9.38 |
dashscope/qwen3.7-max-2026-05-20 |
131k | $3.12 | $9.38 |
dashscope/qwen3.7-max-2026-06-08 |
131k | $3.12 | $9.38 |
dashscope/qwen3.7-max-preview |
131k | $3.12 | $9.38 |
dashscope/qwen3.7-plus |
992k | $1.50 | $6.00 |
dashscope/qwen3.7-plus-2026-05-26 |
131k | $1.50 | $6.00 |
dashscope/qwen3.8-2.4t-a95b |
1M | $2.50 | $7.50 |
dashscope/qwen3.8-27b |
1M | $0.62 | $3.75 |
dashscope/qwen3.8-flash |
1M | $0.19 | $0.59 |
dashscope/qwen3.8-max |
1M | $2.50 | $7.50 |
dashscope/qwen3.8-max-0902 |
262k? | $2.50 | $7.50 |
dashscope/qwq-plus |
98k | $1.00 | $3.00 |
dashscope/qwq-plus-2025-03-05 |
131k | $1.00 | $3.00 |
Z.ai
| Model | Context | Input ($/M) | Cached in ($/M) | Output ($/M) |
|---|---|---|---|---|
zai/glm-4.5 |
128k | $0.75 | — | $2.75 |
zai/glm-4.5-air |
128k | $0.25 | — | $1.38 |
zai/glm-4.6 |
205k | $0.75 | $0.14 | $2.75 |
zai/glm-4.7 |
200k | $0.75 | — | $2.75 |
zai/glm-5 |
200k | $1.25 | — | $4.00 |
zai/glm-5-turbo |
203k | $1.50 | — | $5.00 |
zai/glm-5.1 |
203k | $1.75 | — | $5.50 |
zai/glm-5.2 |
1M | $1.75 | — | $5.50 |
zai/glm-5.3 |
1M | $1.75 | — | $5.50 |
zai/glm-5.3-flash |
1M | $0.19 | — | $0.62 |
DeepSeek
| Model | Context | Input ($/M) | Output ($/M) |
|---|---|---|---|
deepseek-v4-flash |
1.04858M | $0.18 | $0.35 |
deepseek-v4-flash-vision-exp |
262k? | $0.18 | $0.35 |
deepseek-v4-pro |
1.04858M | $0.54 | $1.09 |
MiniMax
| Model | Context | Input ($/M) | Cached in ($/M) | Output ($/M) |
|---|---|---|---|---|
minimax/MiniMax-M2 |
205k | $0.38 | $0.04 | $1.50 |
minimax/MiniMax-M2.1 |
1M | $0.38 | $0.04 | $1.50 |
minimax/MiniMax-M2.1-highspeed |
200k? | $0.75 | — | $3.00 |
minimax/MiniMax-M2.5 |
205k | $0.38 | $0.04 | $1.50 |
minimax/MiniMax-M2.5-highspeed |
205k | $0.75 | — | $3.00 |
minimax/MiniMax-M2.7 |
205k | $0.38 | — | $1.50 |
minimax/MiniMax-M2.7-highspeed |
205k | $0.75 | — | $3.00 |
minimax/MiniMax-M3 |
1M | $0.38 | $0.15 | $1.50 |
คิดเงินตามหน่วยผลลัพธ์ ไม่ใช่ตาม token
| Model | $/1k ตัวอักษร |
|---|---|
minimax/speech-02-hd |
$0.19 |
xAI
| Model | Context | Input ($/M) | Output ($/M) |
|---|---|---|---|
xai/grok-4.20-0309-non-reasoning |
1M | $1.56 | $3.12 |
xai/grok-4.20-0309-reasoning |
2M | $1.56 | $3.12 |
xai/grok-4.20-multi-agent-0309 |
1M | $1.56 | $3.12 |
xai/grok-4.3 |
1M | $1.56 | $3.12 |
xai/grok-4.5 |
500k | $2.50 | $7.50 |
xai/grok-4.6 |
500k | $2.50 | $7.50 |
xai/grok-build-0.1 |
256k | $1.25 | $2.50 |
Moonshot
| Model | Context | Input ($/M) | Output ($/M) |
|---|---|---|---|
moonshot/kimi-k2.6 |
262k | $1.19 | $5.00 |
moonshot/kimi-k2.7-code |
262k | $1.19 | $5.00 |
moonshot/kimi-k2.7-code-highspeed |
262k | $2.38 | $10.00 |
moonshot/kimi-k3 |
1.04858M | $3.75 | $18.75 |
Groq
| Model | Context | Input ($/M) | Output ($/M) |
|---|---|---|---|
groq/openai/gpt-oss-120b |
131k | $0.19 | $0.75 |
groq/openai/gpt-oss-20b |
131k | $0.09 | $0.38 |
groq/qwen/qwen3.6-27b |
131k | $0.75 | $3.75 |
groq/qwen/qwen3.8-27b |
131k | $1.00 | $5.00 |
LTX
คิดเงินตามหน่วยผลลัพธ์ ไม่ใช่ตาม token
| Model | ความละเอียด | $/วินาที |
|---|---|---|
ltx/ltx-2-3-fast |
1080P | $0.07 |
ltx/ltx-2-3-pro |
1080P | $0.10 |
ltx/ltx-2-3-pro-audio |
1080P | $0.12 |
ltx/ltx-2-5-fast |
1080P | $0.16 |
ltx/ltx-2-5-pro |
1080P | $0.21 |
ElevenLabs
คิดเงินตามหน่วยผลลัพธ์ ไม่ใช่ตาม token
| Model | $/1k ตัวอักษร |
|---|---|
elevenlabs/eleven_flash_v2_5 |
$0.06 |
elevenlabs/eleven_multilingual_v2 |
$0.12 |
elevenlabs/eleven_turbo_v2_5 |
$0.06 |
elevenlabs/eleven_v3 |
$0.12 |
Kie
คิดเงินตามหน่วยผลลัพธ์ ไม่ใช่ตาม token
| Model | ความละเอียด | $/วินาที |
|---|---|---|
kie/bytedance/seedance-2 |
1080P | $0.19 |
kie/bytedance/seedance-2 |
720P | $0.07 |
kie/bytedance/seedance-2-fast |
1080P | $0.19 |
kie/bytedance/seedance-2-fast |
720P | $0.07 |
kie/grok-imagine/image-to-video |
720P | $0.02 |
kie/grok-imagine/text-to-video |
720P | $0.02 |
kie/veo3 |
720P | $0.31 |
kie/veo3_fast |
720P | $0.06 |
Row ที่ inactive / deprecated
| โมเดล | เหตุผล |
|---|---|
anthropic/claude-haiku-4-5 |
ถูกแทนที่ด้วย alias ที่ระบุวันที่ claude-haiku-4-5-20251001 — /v1/models ของ Anthropic ไม่ return alias เปล่าแล้ว เก็บ row ไว้ด้วย active=FALSE เพื่อให้ usage row เก่า ๆ ยัง join ได้ |
หมายเหตุเรื่องความถูกต้องของราคา
ราคากับ context window เชื่อถือได้ไม่เท่ากัน ตารางด้านบนจึงแยกให้เห็น:
| ข้อมูล | เชื่อได้แค่ไหน |
|---|---|
| ราคา | ทุกแถวสาวกลับไปยังแหล่งที่ประกาศไว้ได้ — หน้าเว็บของ provider เองหรือ feed ของ LiteLLM ถ้าแถวไหนสาวไม่ถึง make audit-pricing จะ fail ตอน release |
context ที่ไม่มี ? |
มีที่มา: API ของ provider, เอกสารของ provider หรือ LiteLLM |
context ที่มี ? |
เป็นค่า default กลาง ๆ ของ provider นั้น — ถือเป็นขอบล่าง ไม่ใช่สเปก คิดเป็นราว 1 ใน 5 ของแถวทั้งหมด กระจุกอยู่ที่ Qwen Cloud กับ DashScope ซึ่ง API ไม่คืนค่า window มาให้ |
openrouter/auto, openrouter/fusion* |
Pass-through — ต้นทุนเปลี่ยนไปตามแต่ละ request จึงไม่มีอัตราตายตัวให้แสดง |
ถ้าพบ price drift ให้รัน:
python3 scripts/refresh-model-catalogue.py --reprice-only # dry-run
python3 scripts/refresh-model-catalogue.py --reprice-only --apply # commit
ปรับ platform markup
ตัวคูณ 1.25× เป็น env var ตัวเดียวบน gateway service:
# thclaws-cloud/docker-compose.yml
gateway:
environment:
THCLAWS_PLATFORM_MARKUP: ${THCLAWS_PLATFORM_MARKUP:-1.25}
เปลี่ยนเป็น 1.50 แล้ว bounce gateway คือวิธีที่เร็วที่สุดในการเพิ่ม
margin แบบ uniform ทั่วทุกโมเดล gateway จะ clamp ค่าใด ๆ ที่ต่ำกว่า
1.0 (รันที่ราคา upstream เป๊ะ ๆ คือขั้นต่ำ — ต่ำกว่านี้คือขาดทุนต่อ
การเรียก) และจะ emit warning
ถ้าอยากตั้งราคาต่างกันต่อโมเดล แก้ row ใน model_pricing ตรง ๆ ได้ —
แต่ระวังว่า refresh script จะ reprice กลับไปเป็น LiteLLM ใน run
--reprice ครั้งถัดไป ทางออกคือตั้ง active=FALSE แล้ว active=TRUE
ด้วยอัตรา custom พร้อมตั้ง --providers scope ให้ไม่รวม provider นั้น
โค้ดที่เกี่ยวข้อง
| เรื่อง | ไฟล์ |
|---|---|
| คำนวณ cost | thclaws-cloud/gateway/src/pricing.rs (cost_cents_with_markup) |
| Parse markup env var | thclaws-cloud/gateway/src/config.rs (platform_markup) |
| จุดเรียก meter | thclaws-cloud/gateway/src/meter.rs |
| Schema ของ pricing table | thclaws-cloud/api/alembic/versions/006_*.py |
| Refresh script | scripts/refresh-model-catalogue.py |