feat(providers): add SambaNova as an OpenAI-compatible model provider

coworker/providers/matrix.py

@@ -109,6 +109,26 @@ class ModelEntry:
     "mistral:mistral-large-latest": ModelEntry(
         "Mistral Large · Mistral", _AGENTIC, 128_000
     ),
+    # -- SambaNova ---------------------------------------------------------------
+    # OpenAI-compatible API (verified against api.sambanova.ai/v1/models 2026-07-30).
+    "sambanova:Meta-Llama-3.3-70B-Instruct": ModelEntry(
+        "Llama 3.3 70B · SambaNova", _AGENTIC, 131_072
+    ),
+    "sambanova:DeepSeek-V3.1": ModelEntry(
+        "DeepSeek V3.1 · SambaNova", _AGENTIC, 131_072
+    ),
+    "sambanova:DeepSeek-V3.2": ModelEntry(
+        "DeepSeek V3.2 · SambaNova", _AGENTIC, 32_768
+    ),
+    "sambanova:MiniMax-M2.7": ModelEntry(
+        "MiniMax M2.7 · SambaNova", _AGENTIC, 196_608
+    ),
+    "sambanova:gemma-4-31B-it": ModelEntry(
+        "Gemma 4 31B · SambaNova", _AGENTIC, 131_072
+    ),
+    "sambanova:gpt-oss-120b": ModelEntry(
+        "GPT-OSS 120B · SambaNova", _AGENTIC, 131_072
+    ),
     # -- resellers (their model namespaces, verbatim) -----------------------------
     "together:thinkingmachines/Inkling": ModelEntry("Inkling · via Together"),
     "together:zai-org/GLM-5.2": ModelEntry("GLM-5.2 · via Together", _AGENTIC, 128_000),
@@ -157,6 +177,16 @@ class ModelEntry:
     "openrouter:meta-llama/llama-4-maverick": ModelEntry(
         "Llama 4 Maverick · via OpenRouter", _AGENTIC, 1_000_000
     ),
+    # -- SambaNova ------------------------------------------------------
+    "sambanova:Meta-Llama/Llama-4-Maverick-17B-128E-Instruct": ModelEntry(
+        "Llama 4 Maverick · SambaNova", _AGENTIC, 1_000_000
+    ),
+    "sambanova:Qwen/Qwen3-72B-Instruct": ModelEntry(
+        "Qwen3 72B · SambaNova", _AGENTIC, 128_000
+    ),
+    "sambanova:deepseek-ai/DeepSeek-V3-0324": ModelEntry(
+        "DeepSeek V3 · SambaNova", _AGENTIC, 128_000
+    ),
     # -- cloud accounts (models running in the user's own AWS/GCP) ----------------
     # Bedrock ids carry a family segment (claude/ → native Anthropic path, other/ →
     # Converse) plus AWS's own `-v<n>:<m>` version suffix. Some regions require the

coworker/providers/registry.py

@@ -519,6 +519,14 @@ def _compat(
         env_key="META_API_KEY",
         endpoint_help="Prefilled with the Meta Model API endpoint (public preview, US-only as of 2026-07).",
     ),
+    _compat(
+        "sambanova",
+        "SambaNova",
+        base_url="https://api.sambanova.ai/v1",
+        recommended_model="Meta-Llama-3.3-70B-Instruct",
+        env_key="SAMBANOVA_API_KEY",
+        endpoint_help="Prefilled with SambaNova's official OpenAI-compatible endpoint.",
+    ),
     # Resellers: many labs' models behind one key, using THEIR model namespaces (the curated
     # ids + display labels live in providers/matrix.py). TODO: add Groq here (+ its matrix
     # rows) once the current provider surface is tested — deliberately deferred to bound

surfaces/gui/src/providers/ProviderSetup.tsx

@@ -32,6 +32,7 @@ export const KEY_HELP: Record<string, { url: string; label: string }> = {
   qwen: { url: "https://modelstudio.console.alibabacloud.com", label: "alibabacloud.com" },
   minimax: { url: "https://platform.minimax.io", label: "platform.minimax.io" },
   xai: { url: "https://console.x.ai", label: "console.x.ai" },
+  sambanova: { url: "https://console.sambanova.cloud", label: "console.sambanova.cloud" },
 };
 
 export type Verify = { state: "idle" | "testing" | "ok" | "error"; msg?: string };

surfaces/gui/src/providers/logos.ts

@@ -22,6 +22,7 @@ import qwen from "./logos/qwen.svg";
 import minimax from "./logos/minimax.svg";
 import xai from "./logos/xai.svg";
 import meta from "./logos/meta.svg";
+import sambanova from "./logos/sambanova.svg";
 
 export const PROVIDER_LOGOS: Record<string, string> = {
   anthropic,
@@ -41,6 +42,7 @@ export const PROVIDER_LOGOS: Record<string, string> = {
   qwen,
   minimax,
   xai,
+  sambanova,
 };
 
 export const PROVIDER_ORDER = [
@@ -61,6 +63,7 @@ export const PROVIDER_ORDER = [
   "qwen",
   "minimax",
   "xai",
+  "sambanova",
 ];
 
 export function providerRank(name: string): number {