feat(llm): add opencode-go provider support

docs/content/docs/(configuration)/config.mdx

@@ -32,6 +32,7 @@ deepseek_key = "env:DEEPSEEK_API_KEY"
 xai_key = "env:XAI_API_KEY"
 mistral_key = "env:MISTRAL_API_KEY"
 opencode_zen_key = "env:OPENCODE_ZEN_API_KEY"
+opencode_go_key = "env:OPENCODE_GO_API_KEY"
 
 # Custom LLM providers (alternative to legacy keys)
 [llm.provider.my_anthropic]
@@ -337,6 +338,7 @@ If you define a custom provider with the same ID as a legacy key, your custom co
 | `xai_key` | string | None | XAI API key (or `env:VAR_NAME`) |
 | `mistral_key` | string | None | Mistral API key (or `env:VAR_NAME`) |
 | `opencode_zen_key` | string | None | OpenCode Zen API key (or `env:VAR_NAME`) |
+| `opencode_go_key` | string | None | OpenCode Go API key (or `env:VAR_NAME`) |
 
 #### Custom Providers
 

docs/content/docs/(deployment)/roadmap.mdx

@@ -18,7 +18,7 @@ The full message-in → LLM → response-out pipeline is wired end-to-end across
 - **Config** — hierarchical TOML with `Config`, `AgentConfig`, `ResolvedAgentConfig`, `Binding`, `MessagingConfig`. File watcher with event filtering and content hash debounce for hot-reload.
 - **Multi-agent** — per-agent database isolation, `Agent` struct bundles all dependencies
 - **Database connections** — SQLite + LanceDB + redb per-agent, migrations for all tables
-- **LLM** — `SpacebotModel` implements Rig's `CompletionModel`, routes through `LlmManager` via HTTP with retries and fallback chains across 11 providers (Anthropic, OpenAI, OpenRouter, Z.ai, Groq, Together, Fireworks, DeepSeek, xAI, Mistral, OpenCode Zen)
+- **LLM** — `SpacebotModel` implements Rig's `CompletionModel`, routes through `LlmManager` via HTTP with retries and fallback chains across 12 providers (Anthropic, OpenAI, OpenRouter, Z.ai, Groq, Together, Fireworks, DeepSeek, xAI, Mistral, OpenCode Zen, OpenCode Go)
 - **Model routing** — `RoutingConfig` with process-type defaults, task overrides, fallback chains
 - **Memory** — full stack: types, SQLite store (CRUD + graph), LanceDB (embeddings + vector + FTS), fastembed, hybrid search (RRF fusion). `memory_type` filter wired end-to-end through SearchConfig. `total_cmp` for safe sorting.
 - **Memory maintenance** — decay + prune implemented

interface/src/api/client.ts

@@ -724,6 +724,7 @@ export interface ProviderStatus {
 	gemini: boolean;
 	ollama: boolean;
 	opencode_zen: boolean;
+	opencode_go: boolean;
 	nvidia: boolean;
 	minimax: boolean;
 	minimax_cn: boolean;

interface/src/components/ModelSelect.tsx

@@ -27,6 +27,7 @@ const PROVIDER_LABELS: Record<string, string> = {
   zhipu: "Z.ai (GLM)",
   ollama: "Ollama",
   "opencode-zen": "OpenCode Zen",
+  "opencode-go": "OpenCode Go",
   minimax: "MiniMax",
   "minimax-cn": "MiniMax CN",
 };
@@ -141,6 +142,7 @@ export function ModelSelect({
     "fireworks",
     "zhipu",
     "opencode-zen",
+    "opencode-go",
     "minimax",
     "minimax-cn",
   ];

interface/src/lib/providerIcons.tsx

@@ -112,6 +112,7 @@ export function ProviderIcon({ provider, className = "text-ink-faint", size = 24
 		"zai-coding-plan": ZAI,
 		ollama: OllamaIcon,
 		"opencode-zen": OpenCodeZenIcon,
+		"opencode-go": OpenCodeZenIcon,
 		nvidia: NvidiaIcon,
 		minimax: Minimax,
 		"minimax-cn": Minimax,

interface/src/routes/Settings.tsx

@@ -86,6 +86,14 @@ const PROVIDERS = [
 		envVar: "OPENCODE_ZEN_API_KEY",
 		defaultModel: "opencode-zen/kimi-k2.5",
 	},
+	{
+		id: "opencode-go",
+		name: "OpenCode Go",
+		description: "Lite OpenCode model catalog and limits",
+		placeholder: "...",
+		envVar: "OPENCODE_GO_API_KEY",
+		defaultModel: "opencode-go/kimi-k2.5",
+	},
 	{
 		id: "anthropic",
 		name: "Anthropic",

src/api/models.rs

@@ -113,6 +113,11 @@ fn direct_provider_mapping(models_dev_id: &str) -> Option<&'static str> {
         "togetherai" => Some("together"),
         "fireworks-ai" => Some("fireworks"),
         "zhipuai" => Some("zhipu"),
+        "opencode" => Some("opencode-zen"),
+        "opencode-go" => Some("opencode-go"),
+        "zai-coding-plan" => Some("zai-coding-plan"),
+        "minimax" => Some("minimax"),
+        "moonshotai" => Some("moonshot"),
         _ => None,
     }
 }
@@ -141,108 +146,7 @@ fn as_openai_chatgpt_model(model: &ModelInfo) -> Option<ModelInfo> {
 /// Models from providers not in models.dev (private/custom endpoints).
 fn extra_models() -> Vec<ModelInfo> {
     vec![
-        ModelInfo {
-            id: "opencode-zen/kimi-k2.5".into(),
-            name: "Kimi K2.5".into(),
-            provider: "opencode-zen".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: true,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "opencode-zen/kimi-k2".into(),
-            name: "Kimi K2".into(),
-            provider: "opencode-zen".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "opencode-zen/kimi-k2-thinking".into(),
-            name: "Kimi K2 Thinking".into(),
-            provider: "opencode-zen".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: true,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "opencode-zen/glm-5".into(),
-            name: "GLM 5".into(),
-            provider: "opencode-zen".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "opencode-zen/minimax-m2.5".into(),
-            name: "MiniMax M2.5".into(),
-            provider: "opencode-zen".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "opencode-zen/qwen3-coder".into(),
-            name: "Qwen3 Coder 480B".into(),
-            provider: "opencode-zen".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "opencode-zen/big-pickle".into(),
-            name: "Big Pickle".into(),
-            provider: "opencode-zen".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        // Z.AI Coding Plan
-        ModelInfo {
-            id: "zai-coding-plan/glm-4.7".into(),
-            name: "GLM 4.7 (Coding)".into(),
-            provider: "zai-coding-plan".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "zai-coding-plan/glm-5".into(),
-            name: "GLM 5 (Coding)".into(),
-            provider: "zai-coding-plan".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        ModelInfo {
-            id: "zai-coding-plan/glm-4.5-air".into(),
-            name: "GLM 4.5 Air (Coding)".into(),
-            provider: "zai-coding-plan".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: false,
-            input_audio: false,
-        },
-        // MiniMax
-        ModelInfo {
-            id: "minimax/MiniMax-M2.5".into(),
-            name: "MiniMax M2.5".into(),
-            provider: "minimax".into(),
-            context_window: Some(200000),
-            tool_call: true,
-            reasoning: true,
-            input_audio: false,
-        },
-        // MiniMax CN
+        // MiniMax CN - China-specific endpoint, not on models.dev
         ModelInfo {
             id: "minimax-cn/MiniMax-M2.5".into(),
             name: "MiniMax M2.5".into(),
@@ -252,16 +156,7 @@ fn extra_models() -> Vec<ModelInfo> {
             reasoning: true,
             input_audio: false,
         },
-        // Moonshot AI (Kimi)
-        ModelInfo {
-            id: "moonshot/kimi-k2.5".into(),
-            name: "Kimi K2.5".into(),
-            provider: "moonshot".into(),
-            context_window: None,
-            tool_call: true,
-            reasoning: true,
-            input_audio: false,
-        },
+        // Moonshot AI (Kimi) - moonshot-v1-8k not on models.dev
         ModelInfo {
             id: "moonshot/moonshot-v1-8k".into(),
             name: "Moonshot V1 8K".into(),
@@ -433,6 +328,9 @@ pub(super) async fn configured_providers(config_path: &std::path::Path) -> Vec<&
     if has_key("opencode_zen_key", "OPENCODE_ZEN_API_KEY") {
         providers.push("opencode-zen");
     }
+    if has_key("opencode_go_key", "OPENCODE_GO_API_KEY") {
+        providers.push("opencode-go");
+    }
     if has_key("minimax_key", "MINIMAX_API_KEY") {
         providers.push("minimax");
     }

src/api/providers.rs

@@ -53,6 +53,7 @@ pub(super) struct ProviderStatus {
     gemini: bool,
     ollama: bool,
     opencode_zen: bool,
+    opencode_go: bool,
     nvidia: bool,
     minimax: bool,
     minimax_cn: bool,
@@ -137,6 +138,7 @@ fn provider_toml_key(provider: &str) -> Option<&'static str> {
         "gemini" => Some("gemini_key"),
         "ollama" => Some("ollama_base_url"),
         "opencode-zen" => Some("opencode_zen_key"),
+        "opencode-go" => Some("opencode_go_key"),
         "nvidia" => Some("nvidia_key"),
         "minimax" => Some("minimax_key"),
         "minimax-cn" => Some("minimax_cn_key"),
@@ -253,6 +255,13 @@ fn build_test_llm_config(provider: &str, credential: &str) -> crate::config::Llm
             name: None,
             use_bearer_auth: false,
         }),
+        "opencode-go" => Some(ProviderConfig {
+            api_type: ApiType::OpenAiCompletions,
+            base_url: "https://opencode.ai/zen/go".to_string(),
+            api_key: credential.to_string(),
+            name: None,
+            use_bearer_auth: false,
+        }),
         "nvidia" => Some(ProviderConfig {
             api_type: ApiType::OpenAiCompletions,
             base_url: "https://integrate.api.nvidia.com".to_string(),
@@ -310,6 +319,7 @@ fn build_test_llm_config(provider: &str, credential: &str) -> crate::config::Llm
         ollama_key: None,
         ollama_base_url: (provider == "ollama").then(|| credential.to_string()),
         opencode_zen_key: (provider == "opencode-zen").then(|| credential.to_string()),
+        opencode_go_key: (provider == "opencode-go").then(|| credential.to_string()),
         nvidia_key: (provider == "nvidia").then(|| credential.to_string()),
         minimax_key: (provider == "minimax").then(|| credential.to_string()),
         minimax_cn_key: (provider == "minimax-cn").then(|| credential.to_string()),
@@ -460,6 +470,7 @@ pub(super) async fn get_providers(
         gemini,
         ollama,
         opencode_zen,
+        opencode_go,
         nvidia,
         minimax,
         minimax_cn,
@@ -502,6 +513,7 @@ pub(super) async fn get_providers(
             has_value("ollama_base_url", "OLLAMA_BASE_URL")
                 || has_value("ollama_key", "OLLAMA_API_KEY"),
             has_value("opencode_zen_key", "OPENCODE_ZEN_API_KEY"),
+            has_value("opencode_go_key", "OPENCODE_GO_API_KEY"),
             has_value("nvidia_key", "NVIDIA_API_KEY"),
             has_value("minimax_key", "MINIMAX_API_KEY"),
             has_value("minimax_cn_key", "MINIMAX_CN_API_KEY"),
@@ -524,6 +536,7 @@ pub(super) async fn get_providers(
             std::env::var("GEMINI_API_KEY").is_ok(),
             std::env::var("OLLAMA_BASE_URL").is_ok() || std::env::var("OLLAMA_API_KEY").is_ok(),
             std::env::var("OPENCODE_ZEN_API_KEY").is_ok(),
+            std::env::var("OPENCODE_GO_API_KEY").is_ok(),
             std::env::var("NVIDIA_API_KEY").is_ok(),
             std::env::var("MINIMAX_API_KEY").is_ok(),
             std::env::var("MINIMAX_CN_API_KEY").is_ok(),
@@ -547,6 +560,7 @@ pub(super) async fn get_providers(
         gemini,
         ollama,
         opencode_zen,
+        opencode_go,
         nvidia,
         minimax,
         minimax_cn,
@@ -567,6 +581,7 @@ pub(super) async fn get_providers(
         || providers.gemini
         || providers.ollama
         || providers.opencode_zen
+        || providers.opencode_go
         || providers.nvidia
         || providers.minimax
         || providers.minimax_cn
@@ -891,7 +906,8 @@ pub(super) async fn update_provider(
 pub(super) async fn test_provider_model(
     Json(request): Json<ProviderModelTestRequest>,
 ) -> Result<Json<ProviderModelTestResponse>, StatusCode> {
-    if provider_toml_key(&request.provider).is_none() {
+    let normalized_provider = request.provider.trim().to_lowercase();
+    if provider_toml_key(&normalized_provider).is_none() {
         return Ok(Json(ProviderModelTestResponse {
             success: false,
             message: format!("Unknown provider: {}", request.provider),
@@ -921,7 +937,7 @@ pub(super) async fn test_provider_model(
         }));
     }
 
-    if !model_matches_provider(&request.provider, &request.model) {
+    if !model_matches_provider(&normalized_provider, &request.model) {
         return Ok(Json(ProviderModelTestResponse {
             success: false,
             message: format!(
@@ -934,7 +950,7 @@ pub(super) async fn test_provider_model(
         }));
     }
 
-    let llm_config = build_test_llm_config(&request.provider, request.api_key.trim());
+    let llm_config = build_test_llm_config(&normalized_provider, request.api_key.trim());
     let llm_manager = match crate::llm::LlmManager::new(llm_config).await {
         Ok(manager) => Arc::new(manager),
         Err(error) => {