Code Generation APIs
Models optimized for programming
Hy4 Flash
Tencent Hy4 Flash 1M-context multimodal model free for 2 weeks on OpenCode Zen with Zero Data Retention.
MiniMax M3
MiniMax M3 via OpenCode Go ($5 first month / $10/mo for $60 monthly quota pool).
Kimi K3
Moonshot Kimi K3 via OpenCode Go ($5 first month / $10/mo for $60 monthly quota pool).
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash via OpenCode Go ($5 first month / $10/mo for $60 monthly quota pool).
Muse Spark 1.3
Meta Muse Spark 1.3 Contributor via OpenCode Go ($5 first month / $10/mo) with near-unlimited coding quota.
MiniMax M3.1 Flash
MiniMax M3.1 Flash native multimodal MoE on NVIDIA NIM — agentic coding & computer use at 40 RPM.
Mistral Large 4 1T
Mistral Large 4 1T MoE (52B active) on NVIDIA NIM with NVFP4 and DSpark speculative decoding — 40 RPM.
Step 5 Preview
StepFun 600B MoE (27B active) multimodal flagship preview on NVIDIA NIM in NVFP4 — 1M context, 40 RPM.
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash 552B MoE (8B active) on NVIDIA NIM — 1M context, multimodal, free preview at 40 RPM.
Ling 3.1 Flash 1019
Access Ling 3.1 Flash 1019 free through Vercel AI Gateway $5 monthly credit and limited-time free tier.
Pixel Canary
Access Pixel Canary free through Vercel AI Gateway $5 monthly credit.
Space Bunny Alpha
Space Bunny Alpha stealth agent router and long-context coding model at zero credit cost on Command Code.