DeepSeek Model Error: max_tokens Capped at 131k Despite Official 384k Limit

locustbaby · ghdev · 2026-08-21

A user encountered an error when using the opencode-go integration with DeepSeek Flash, stating that a maxtokens setting of 384,000 exceeds the supported limit of 131,072 completion tokens. This discrepancy suggests a constraint within the OpenCode implementation or its upstream connection, despite DeepSeek's official documentation claiming a higher limit.

Original post →

More from Infra

• DSCO Router Launches Unified Gateway for Multi-Model Routing with BYOK Support — arthurcolle · 2026-08-24
• Open Source RobotSoul: Persistent Identity for Agents After Context Resets — robauto-dot-ai · 2026-08-24
• Offloading MoE models to RAM causes slow prefill speeds — former_farmer · 2026-08-24
• Etched Raises $1B Led by Jane Street to Validate Architecture-Agnostic AI Chips — TheTuringPost · 2026-08-24
• ConvRot Quant joins llama-cpp: Q6 accuracy nears Q8 quality — giveen · 2026-08-24
• LifeOS: A Local, Voice-Driven Personal Organizer — Extension-Bid-639 · 2026-08-24

Infra channel →