# v0.4.66 (2026-05-29)

## Features
- Add Qoder provider: device-flow OAuth, COSY signing, WAF-bypass body encoding, live model catalog, dashboard quota tracker, 11 models (#1372)
- Add new models: Claude Opus 4.8 (Claude Code), GPT 5.4 Mini (Codex)

## Fixes
- DeepSeek thinking mode: echo `reasoning_content` back on follow-up/tool-call turns so OpenCode-free and custom providers no longer 400 with "reasoning_content must be passed back" (#1543)
- Reasoning injector: match deepseek/kimi model ids case-insensitively (covers custom providers using capitalized model names)
- OpenCode suggested-models: include free models without the `-free` suffix, e.g. `big-pickle` (#1535)

## Improvements
- Codex: trim sunset models, keep gpt-5.5 / gpt-5.4 / gpt-5.3-codex family, add gpt-5.4-mini
- volcengine-ark: refresh model list (add DeepSeek-V4-Flash/Pro, drop EOL entries)
- Lower stream stall timeout 35s → 30s for faster hang detection

open-sse/config/runtimeConfig.js

@@ -32,7 +32,7 @@ export const MEMORY_CONFIG = {
 };
 
 // Stream stall timeout: abort if no chunk received within this duration
-export const STREAM_STALL_TIMEOUT_MS = 35 * 1000;
+export const STREAM_STALL_TIMEOUT_MS = 30 * 1000;
 
 // Fetch connect timeout: abort if upstream doesn't return response headers within this duration
 export const FETCH_CONNECT_TIMEOUT_MS = 20 * 1000;

open-sse/executors/opencode.js

@@ -1,14 +1,19 @@
 import { BaseExecutor } from "./base.js";
 import { PROVIDERS } from "../config/providers.js";
+import { injectReasoningContent } from "../utils/reasoningContentInjector.js";
 
 // Models that use /zen/v1/messages (claude format)
-const MESSAGES_MODELS = new Set(["big-pickle"]);
+const MESSAGES_MODELS = new Set();
 
 export class OpenCodeExecutor extends BaseExecutor {
   constructor() {
     super("opencode", PROVIDERS.opencode);
   }
 
+  transformRequest(model, body) {
+    return injectReasoningContent({ provider: this.provider, model, body });
+  }
+
   buildUrl(model) {
     const base = "https://opencode.ai";
     return MESSAGES_MODELS.has(model)

open-sse/utils/reasoningContentInjector.js

@@ -11,8 +11,8 @@ const PROVIDER_RULES = {
 
 // Model-level rules: matched by predicate against model id
 const MODEL_RULES = [
-  { match: m => m?.startsWith?.("kimi-"), scope: "toolCalls" },
-  { match: m => m?.startsWith?.("deepseek-"), scope: "all" }
+  { match: m => /^kimi-/i.test(m || ""), scope: "toolCalls" },
+  { match: m => /deepseek/i.test(m || ""), scope: "all" }
 ];
 
 const DEEPSEEK_V4_PRO = "deepseek-v4-pro";