Models & Pricing
Every model GWC routes has provider-specific pricing. Compare the main chat models and media generators from Anthropic, OpenAI, Google Gemini, xAI Grok, Alibaba Qwen, Z.AI GLM, Kimi, BytePlus, and DeepSeek below. Chat prices below are in USD per 1M fresh input or output tokens, with the 1.15× markup already applied. Scheduled rates show off-peak–peak ranges; their UTC windows appear beside each model. BYOK uses your provider's own billing and rates.
18 models·8 providers·Markup 15%
How billing works
- Basic (Free): GWC Plan only ($1/mo allowance ± top-up balance). On Basic the monthly allowance funds DeepSeek V4 Flash only — any other model on this page is payable from top-up balance, BYOK or Login-OA. Standard+ ($19+), or Basic with the Connect Pass: two funding sources — BYOK (your own vendor keys; free Login-OA sign-in is the primary credential) and GWC Plan (monthly allowance ± top-up balance).
- GWC Plan: billed at 1.15× vendor cost (prices shown below).
- BYOK (incl. free Login-OA): bypasses our metering — you pay the vendor directly at vendor cost (divide by 1.15).
Anthropic — Claude
3 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
Fable 5.1claude-fable-5-1 | deepest reasoning | $11.50 | $57.50 |
Opus 5claude-opus-5 | most capable Claude · default | $5.75 | $28.75 |
Sonnet 5claude-sonnet-5 | balanced flagship | $3.45 | $17.25 |
OpenAI — GPT
3 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
GPT-5.6 Solgpt-5.6-sol | 1M ctx flagship | $5.75 | $34.50 |
GPT-5.6 Terragpt-5.6-terra | half-cost flagship | $2.875 | $17.25 |
GPT-5.6 Lunagpt-5.6-luna | fast & affordable | $1.15 | $6.90 |
Google — Gemini
3 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
Gemini 3.1 Progemini-3.1-pro-preview | 1M ctx | $2.30 | $13.80 |
Gemini 3.7 Flashgemini-3.7-flash | newest Flash | $0.8625 | $4.3125 |
Gemini 3.6 Flashgemini-3.6-flash | previous Flash generation | $0.8625 | $4.3125 |
xAI — Grok
1 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
Grok 4.6grok-4.6 | frontier coder | $2.30 | $6.90 |
Alibaba — Qwen
2 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
Qwen3.8-Maxqwen3.8-max | flagship · 1M ctx | $2.30 | $6.90 |
Qwen3.7-Flashqwen3.7-flash | very cheap · 1M ctx | $0.115 | $0.46 |
Z.AI — GLM
3 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
GLM-5.3glm-5.3 | 1M ctx · 128K output · always-on reasoning | $1.61 | $5.06 |
GLM-5.3-Flashglm-5.3-flash | 1M ctx · 18B active · cheapest agentic tier | $0.1725 | $0.575 |
GLM-5.2glm-5.2 | 1M ctx · previous generation | $1.61 | $5.06 |
Kimi
1 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
Kimi K3kimi-k3 | flagship · 1M ctx · BYOK | $3.45 | $17.25 |
DeepSeek
2 models
| Model | Reference | Input / 1M tok | Output / 1M tok |
|---|---|---|---|
DeepSeek-V4-Prodeepseek-v4-pro | strong coder, big context Off-peak–peak rates. Peak: 01:00–04:00 and 06:00–10:00 UTC daily (end times excluded). Lower rate at all other times. Effective 2026-08-14. | $0.759–$1.518 | $2.277–$4.554 |
DeepSeek-V4-Flashdeepseek-v4-flash | fast & economical (floor) Off-peak–peak rates. Peak: 01:00–04:00 and 06:00–10:00 UTC daily (end times excluded). Lower rate at all other times. Effective 2026-08-14. | $0.253–$0.506 | $0.759–$1.518 |
Media Generation Rates
Image, video, and 3D generation pricing — Seedream, Imagen, Wan, Veo, Seedance, Hyper3D. Prices are per-request with the 1.15× markup already applied. BYOK users pay vendor cost directly (divide listed prices by 1.15).
47 models·3 categories·Markup 15%
How media billing works
- Image generation: billed per image output. Price covers one generation request.
- Video generation: Veo and Wan are billed per video (5s 720p baseline). Seedance is token-based (per 1M tokens consumed).
- 3D: billed per 1M tokens consumed during generation.
- Basic (Free): GWC Plan only ($1/mo allowance ± top-up balance). Standard+ (or Basic with the Connect Pass): two funding sources — BYOK (your own vendor keys; free Login-OA sign-in is the primary credential) and GWC Plan (monthly allowance ± top-up balance). Media generation available to all tiers.
Image Generation
24 models
| Model | Reference | Price | Unit |
|---|---|---|---|
Seedream 5.0 Liteseedream-5-0-lite | latest lite | $0.040 | per image |
Seedream 4.5seedream-4-5 | flagship | $0.046 | per image |
Seedream 4.0seedream-4-0 | standard | $0.035 | per image |
Seedream 3.0seedream-3-0-t2i | text-to-image | $0.035 | per image |
SeedEdit 3.0seededit-3-0-i2i | image-to-image | $0.035 | per image |
Imagen 4.0 Ultraimagen-4.0-ultra-generate-001 | highest quality | $0.092 | per image |
Imagen 4.0imagen-4.0-generate-001 | standard | $0.046 | per image |
Imagen 4.0 Fastimagen-4.0-fast-generate-001 | fastest | $0.023 | per image |
Qwen Image 2.0 Proqwen-image-2.0-pro | best quality | $0.086 | per image |
Qwen Image 2.0qwen-image-2.0 | standard | $0.040 | per image |
Qwen Image Maxqwen-image-max | max quality | $0.086 | per image |
Z Image Turboz-image-turbo | fast | $0.035 | per image |
Qwen Image Plusqwen-image-plus | standard | $0.035 | per image |
Qwen Imageqwen-image | basic | $0.040 | per image |
Wan 2.7 Image Prowan2.7-image-pro | latest pro | $0.046 | per image |
Wan 2.7 Imagewan2.7-image | latest standard | $0.033 | per image |
Wan 2.6 Imagewan2.6-image | 2.6 standard | $0.033 | per image |
Wan 2.5 T2Iwan2.5-t2i-preview | 2.5 preview | $0.035 | per image |
Wan 2.2 T2I Pluswan2.2-t2i-plus | 2.2 plus | $0.029 | per image |
Wan 2.2 T2I Flashwan2.2-t2i-flash | 2.2 fast | $0.017 | per image |
Wan 2.1 T2I Pluswan2.1-t2i-plus | 2.1 plus | $0.029 | per image |
Wan 2.1 T2I Turbowan2.1-t2i-turbo | 2.1 fast | $0.017 | per image |
Qwen Image Edit Plusqwen-image-edit-plus | editing | $0.035 | per image |
Qwen Image Editqwen-image-edit | basic editing | $0.052 | per image |
Video Generation
21 models
| Model | Reference | Price | Unit |
|---|---|---|---|
Veo 3.1veo-3.1-generate-preview | latest flagship | $0.403 | per video |
Veo 3.1 Fastveo-3.1-fast-generate-preview | fast | $0.172 | per video |
Veo 3.1 Liteveo-3.1-lite-generate-preview | lite | $0.115 | per video |
Veo 3.0veo-3.0-generate-001 | 3.0 standard | $0.403 | per video |
Veo 3.0 Fastveo-3.0-fast-generate-001 | 3.0 fast | $0.172 | per video |
Veo 2.0veo-2.0-generate-001 | 2.0 standard | $0.287 | per video |
Seedance 1.5 Proseedance-1-5-pro | latest flagship | $2.76 | per 1M tokens |
Seedance 1.0 Proseedance-1-0-pro | flagship | $2.88 | per 1M tokens |
Seedance 1.0 Pro Fastseedance-1-0-pro-fast | pro fast | $1.15 | per 1M tokens |
Seedance 1.0 Lite T2Vseedance-1-0-lite-t2v | text-to-video | $2.07 | per 1M tokens |
Seedance 1.0 Lite I2Vseedance-1-0-lite-i2v | image-to-video | $2.07 | per 1M tokens |
Wan 2.7 T2Vwan2.7-t2v | latest text-to-video | $0.494 | per video |
Wan 2.7 VideoEditwan2.7-videoedit | latest video editing | $0.575 | per video |
Wan 2.6 T2Vwan2.6-t2v | 2.6 text-to-video | $0.494 | per video |
Wan 2.6 I2Vwan2.6-i2v | 2.6 image-to-video | $0.494 | per video |
Wan 2.6 I2V Flashwan2.6-i2v-flash | 2.6 i2v fast | $0.494 | per video |
Wan 2.5 T2Vwan2.5-t2v-preview | 2.5 preview | $0.494 | per video |
Wan 2.2 T2V Pluswan2.2-t2v-plus | 2.2 plus | $0.494 | per video |
Wan 2.2 I2V Flashwan2.2-i2v-flash | 2.2 fast | $0.207 | per video |
Wan 2.1 T2V Turbowan2.1-t2v-turbo | 2.1 fast | $0.287 | per video |
Wan 2.1 T2V Pluswan2.1-t2v-plus | 2.1 plus | $0.575 | per video |
3D Generation
2 models
| Model | Reference | Price | Unit |
|---|---|---|---|
Hyper3D Gen 2hyper3d-gen2 | Deemos | $15.29 | per 1M tokens |
Hitem3D 2.0hitem3d-2-0 | Math Magic HK | $23.00 | per 1M tokens |
For agents: fetch live
If you're building an agent that uses GWC, fetch the live catalogue at https://hk.gwc-corp.com/api/v1/catalogue rather than hard-coding model IDs or rates. We add and rename models often; the endpoint returns a 5-minute cache header.