GA
Global API
DocsGet Started

Model List

Supported Models

We support **180+ models** across 10+ AI providers — including our exclusive **GA Series** models. Free models available for zero-cost startup.

✅ All model IDs below have been tested and verified against our One-API backend. Use them exactly as shown.

⭐ GA Series (Flagship)

Model IDCategoryPricingDescription
ga-economyGA-Economy0.5x creditsGlobal API's cost-optimized model. Outperforms 95% of models at half the price.
ga-standardGA-Standard1.0x creditsOur flagship model. Rivals top-tier models at a fraction of the cost.
ga-expressGA-Express1.5x creditsEnterprise speed tier. Ultra-fast responses.

DeepSeek Series 🚀

Model IDCategoryContextNotes
deepseek-chatV4 Flash1M**Best value** — unified chat + reasoning + coding
deepseek-ai/DeepSeek-V4-FlashV4 Flash (alt)1MAlternative route via SiliconFlow
deepseek-v4-flashV4 Flash (alt)1MAlternative route via DeepSeek
deepseek-v4-proV4 Pro1MPremium reasoning (discounted)
deepseek-coderCode specialist—Code generation
deepseek-ai/DeepSeek-V3.2V3.2128KReliable general-purpose
deepseek-v3V3128KSolid baseline
deepseek-ai/DeepSeek-V3V3 (SF)128KSame via SiliconFlow
deepseek-ai/DeepSeek-V3.1-TerminusV3.1128KFast variant
deepseek-reasonerR1128K**Advanced STEM reasoning**
deepseek-ai/DeepSeek-R1R1 (SF)128KSame via SiliconFlow
deepseek-r1R1 (alt)128KAlternative endpoint
deepseek-ai/DeepSeek-R1-0528-Qwen3-8BR1 Lite 8B32KDistilled, low cost
deepseek-ai/DeepSeek-OCROCR—Text extraction from images
deepseek-v3-2-250121V3.2 date128KDated version
deepseek-r1-250121R1 date128KDated version

Qwen / Alibaba Series

Model IDSizeContextBest For
Qwen/Qwen3.6-35B-A3B35B MoE128K**Latest flagship** — efficient
Qwen/Qwen3.6-27B27B128KGen 3.6 standard
Qwen/Qwen3.5-397B-A17B397B MoE128K**Highest intelligence**
Qwen/Qwen3.5-122B-A10B122B MoE128KVery high intelligence
Qwen/Qwen3.5-35B-A3B35B MoE128KBalanced quality
Qwen/Qwen3.5-27B27B128KStandard
Qwen/Qwen3.5-9B9B128KLightweight
Qwen/Qwen3.5-4B4B128KUltra-light
Qwen/Qwen3-32B32B128KGen 3 flagship
Qwen/Qwen3-14B14B128KMid-range
Qwen/Qwen3-8B8B128K**Free** — zero cost
Qwen/Qwen3-235B-A22B-Instruct-2507235B MoE128KMassive MoE
Qwen/Qwen3-30B-A3B-Instruct-250730B MoE128KInstruct tuned
Qwen/Qwen3-30B-A3B-Thinking-250730B MoE128KThinking mode
QwQ-32B32B128K**Qwen reasoning model**
qwen-turbo——Budget (Qwen API)
qwen-plus——Mid (Qwen API)
qwen-max——Premium (Qwen API)

Kimi / Moonshot Series

Model IDContextBest For
kimi-k2.5128K**Latest flagship (Kimi API)**
kimi-k2.6128K**Newest upgrade (Kimi API)**
moonshotai/Kimi-K2-Thinking128KDeep reasoning (SF)
moonshotai/Kimi-K2-Instruct-0905128KInstruct (SF)
moonshot-v1-auto—Auto-select
moonshot-v1-8k8KBudget short context

GLM / Zhipu AI Series

Model IDSizeContextBest For
THUDM/GLM-4-32B-041432B128K**GLM flagship**
THUDM/GLM-Z1-32B-041432B128KGLM reasoning
THUDM/GLM-4-9B-04149B128K**Free** — basic GLM
THUDM/GLM-Z1-9B-04149B128K**Free** — GLM reasoning
THUDM/GLM-4.1V-9B-Thinking9B8K**Free** — vision + thinking

Other Models

Model IDSize/ContextBest For
doubao-seed-2-0-pro-26021564KByteDance premium
MiniMaxAI/MiniMax-M2.51M context**Longest context** — full document analysis
stepfun-ai/Step-3.5-FlashFast**Budget reasoning** — fast + cheap
tencent/Hunyuan-A13B-Instruct13BTencent flagship
inclusionAI/Ring-flash-2.0—Inclusion AI
inclusionAI/Ling-flash-2.0—Inclusion AI alt

Quick Reference

You WantUse This Model ID
Best overall valuedeepseek-chat
DeepSeek V4 Pro (premium)deepseek-v4-pro
DeepSeek R1 (reasoning)deepseek-reasoner
Latest Qwen flagshipQwen/Qwen3.6-35B-A3B
Largest Qwen (397B MoE)Qwen/Qwen3.5-397B-A17B
Qwen reasoningQwQ-32B
Latest Kimikimi-k2.5 or kimi-k2.6
GLM flagshipTHUDM/GLM-4-32B-0414
Free modelQwen/Qwen3-8B or THUDM/GLM-4-9B-0414
Ultra-long context (1M)MiniMaxAI/MiniMax-M2.5
ByteDance premiumdoubao-seed-2-0-pro-260215
Get Started Free →