feat(cli): add Featherless (featherless.ai) as a built-in provider

.env.example

@@ -70,6 +70,15 @@
 # ARCEEAI_API_KEY=
 # ARCEE_BASE_URL=                                 # Override default base URL
 
+# =============================================================================
+# LLM PROVIDER (Featherless)
+# =============================================================================
+# Featherless serves thousands of open-source models (Hugging Face repo IDs)
+# via an OpenAI-compatible API — e.g. zai-org/GLM-5.2, meta-llama/Llama-3.1-8B-Instruct
+# Get a Featherless key at: https://featherless.ai/
+# FEATHERLESS_API_KEY=
+# FEATHERLESS_BASE_URL=                            # Override default base URL
+
 # =============================================================================
 # LLM PROVIDER (MiniMax)
 # =============================================================================

agent/model_metadata.py

@@ -53,6 +53,7 @@ def _resolve_requests_verify() -> bool | str:
     "xiaomi",
     "arcee",
     "gmi",
+    "featherless",
     "tencent-tokenhub",
     "custom", "local",
     # Common aliases
@@ -65,6 +66,7 @@ def _resolve_requests_verify() -> bool | str:
     "tencent", "tokenhub", "tencent-cloud", "tencentmaas",
     "arcee-ai", "arceeai",
     "gmi-cloud", "gmicloud",
+    "featherless-ai", "featherlessai",
     "xai", "x-ai", "x.ai", "grok",
     "nvidia", "nim", "nvidia-nim", "nemotron",
     "qwen-portal", "novita-ai", "novitaai",
@@ -420,6 +422,7 @@ def _is_custom_endpoint(base_url: str) -> bool:
     "api.stepfun.ai": "stepfun",
     "api.stepfun.com": "stepfun",
     "api.arcee.ai": "arcee",
+    "api.featherless.ai": "featherless",
     "api.minimax": "minimax",
     "dashscope.aliyuncs.com": "alibaba",
     "dashscope-intl.aliyuncs.com": "alibaba",
@@ -1635,7 +1638,7 @@ def get_model_context_length(
           cache fallback with suffix/version normalisation.  Only
           portal-derived values are persisted to disk.
        c. Codex OAuth /models probe
-       d. GMI /models endpoint
+       d. GMI / Featherless /models endpoint (provider-served context_length)
        e. Ollama native /api/show probe (any base_url, provider-agnostic)
        f. models.dev registry lookup (with :cloud/-cloud suffix fallback)
     6. OpenRouter live API metadata (Kimi-family 32k guard)
@@ -1882,6 +1885,15 @@ def get_model_context_length(
         ctx = _resolve_endpoint_context_length(model, base_url, api_key=api_key)
         if ctx is not None:
             return ctx
+    if effective_provider == "featherless" and base_url:
+        # Featherless serves each model at a provider-specific context_length
+        # (e.g. zai-org/GLM-5.2 at 256K, not the model's native 1M; an 8B Llama
+        # at 32K) exposed via /v1/models. Prefer that authoritative endpoint
+        # value over models.dev and the hardcoded native-context fallback so
+        # token budgeting matches what the endpoint will actually accept.
+        ctx = _resolve_endpoint_context_length(model, base_url, api_key=api_key)
+        if ctx is not None:
+            return ctx
     # 5e. Ollama native /api/show probe — runs for ANY provider with a
     # base_url, not just ollama-cloud.  Ollama-compatible servers expose
     # this endpoint regardless of hostname (local Ollama, Ollama Cloud,

apps/desktop/src/app/settings/constants.ts

@@ -191,6 +191,13 @@ export const PROVIDER_GROUPS: ProviderPrefix[] = [
     priority: 20
   },
   { prefix: 'ARCEE_', name: 'Arcee AI', priority: 20 },
+  {
+    prefix: 'FEATHERLESS_',
+    name: 'Featherless',
+    description: 'Thousands of open-source models, serverless',
+    docsUrl: 'https://featherless.ai/',
+    priority: 20
+  },
   {
     prefix: 'GMI_',
     name: 'GMI Cloud',

cli-config.yaml.example

@@ -27,6 +27,7 @@ model:
   #   "nvidia"       - NVIDIA NIM / build.nvidia.com (requires: NVIDIA_API_KEY)
   #   "xiaomi"       - Xiaomi MiMo (requires: XIAOMI_API_KEY)
   #   "arcee"        - Arcee AI Trinity models (requires: ARCEEAI_API_KEY)
+  #   "featherless"  - Featherless open-source models (requires: FEATHERLESS_API_KEY)
   #   "ollama-cloud" - Ollama Cloud (requires: OLLAMA_API_KEY — https://ollama.com/settings)
   #   "kilocode"     - KiloCode gateway (requires: KILOCODE_API_KEY)
   #   "azure-foundry" - Microsoft Foundry / Azure OpenAI (API key or Entra ID)

hermes_cli/auth.py

@@ -284,6 +284,14 @@ class ProviderConfig:
         api_key_env_vars=("ARCEEAI_API_KEY",),
         base_url_env_var="ARCEE_BASE_URL",
     ),
+    "featherless": ProviderConfig(
+        id="featherless",
+        name="Featherless",
+        auth_type="api_key",
+        inference_base_url="https://api.featherless.ai/v1",
+        api_key_env_vars=("FEATHERLESS_API_KEY",),
+        base_url_env_var="FEATHERLESS_BASE_URL",
+    ),
     "gmi": ProviderConfig(
         id="gmi",
         name="GMI Cloud",
@@ -1518,6 +1526,7 @@ def resolve_provider(
         "kimi-cn": "kimi-coding-cn", "moonshot-cn": "kimi-coding-cn",
         "step": "stepfun", "stepfun-coding-plan": "stepfun",
         "arcee-ai": "arcee", "arceeai": "arcee",
+        "featherless-ai": "featherless", "featherlessai": "featherless",
         "gmi-cloud": "gmi", "gmicloud": "gmi",
         "minimax-china": "minimax-cn", "minimax_cn": "minimax-cn",
         "minimax-portal": "minimax-oauth", "minimax-global": "minimax-oauth", "minimax_oauth": "minimax-oauth",

hermes_cli/config.py

@@ -2784,6 +2784,22 @@ def _ensure_hermes_home_managed(home: Path):
         "category": "provider",
         "advanced": True,
     },
+    "FEATHERLESS_API_KEY": {
+        "description": "Featherless API key",
+        "prompt": "Featherless API key",
+        "url": "https://featherless.ai/",
+        "password": True,
+        "category": "provider",
+        "advanced": True,
+    },
+    "FEATHERLESS_BASE_URL": {
+        "description": "Featherless base URL override",
+        "prompt": "Featherless base URL (leave empty for default)",
+        "url": None,
+        "password": False,
+        "category": "provider",
+        "advanced": True,
+    },
     "GMI_API_KEY": {
         "description": "GMI Cloud API key",
         "prompt": "GMI Cloud API key",

hermes_cli/doctor.py

@@ -41,6 +41,7 @@
     "KIMI_API_KEY",
     "KIMI_CN_API_KEY",
     "GMI_API_KEY",
+    "FEATHERLESS_API_KEY",
     "MINIMAX_API_KEY",
     "MINIMAX_CN_API_KEY",
     "KILOCODE_API_KEY",
@@ -383,6 +384,7 @@ def _build_apikey_providers_list() -> list:
         ("StepFun Step Plan", ("STEPFUN_API_KEY",),                          "https://api.stepfun.ai/step_plan/v1/models", "STEPFUN_BASE_URL", True),
         ("Kimi / Moonshot (China)", ("KIMI_CN_API_KEY",),                    "https://api.moonshot.cn/v1/models",   None, True),
         ("Arcee AI",         ("ARCEEAI_API_KEY",),                           "https://api.arcee.ai/api/v1/models",  "ARCEE_BASE_URL", True),
+        ("Featherless",      ("FEATHERLESS_API_KEY",),                       "https://api.featherless.ai/v1/models", "FEATHERLESS_BASE_URL", True),
         ("GMI Cloud",        ("GMI_API_KEY",),                               "https://api.gmi-serving.com/v1/models", "GMI_BASE_URL", True),
         ("DeepSeek",         ("DEEPSEEK_API_KEY",),                          "https://api.deepseek.com/v1/models",  "DEEPSEEK_BASE_URL", True),
         ("Hugging Face",     ("HF_TOKEN",),                                  "https://router.huggingface.co/v1/models", "HF_BASE_URL", True),
@@ -404,7 +406,7 @@ def _build_apikey_providers_list() -> list:
     _name_to_canonical = {
         "Z.AI / GLM": "zai", "Kimi / Moonshot": "kimi-coding",
         "StepFun Step Plan": "stepfun", "Kimi / Moonshot (China)": "kimi-coding-cn",
-        "Arcee AI": "arcee", "GMI Cloud": "gmi", "DeepSeek": "deepseek",
+        "Arcee AI": "arcee", "Featherless": "featherless", "GMI Cloud": "gmi", "DeepSeek": "deepseek",
         "Hugging Face": "huggingface", "NVIDIA NIM": "nvidia",
         "Alibaba/DashScope": "alibaba", "MiniMax": "minimax",
         "MiniMax (China)": "minimax-cn",

hermes_cli/main.py

@@ -3040,6 +3040,7 @@ def _active_custom_key_from_base_url() -> str:
         "huggingface",
         "xiaomi",
         "arcee",
+        "featherless",
         "gmi",
         "nvidia",
         "ollama-cloud",

hermes_cli/models.py

@@ -381,6 +381,14 @@ def _xai_curated_models() -> list[str]:
         "trinity-large-preview",
         "trinity-mini",
     ],
+    "featherless": [
+        "zai-org/GLM-5.2",
+        "zai-org/GLM-4.7",
+        "moonshotai/Kimi-K2-Thinking",
+        "deepseek-ai/DeepSeek-V3.2",
+        "Qwen/Qwen3.5-397B-A17B",
+        "meta-llama/Llama-3.1-8B-Instruct",
+    ],
     "gmi": [
         "zai-org/GLM-5.1-FP8",
         "deepseek-ai/DeepSeek-V3.2",
@@ -1040,6 +1048,7 @@ class ProviderEntry(NamedTuple):
     ProviderEntry("minimax-cn",     "MiniMax (China)",          "MiniMax China (Domestic direct API)"),
     ProviderEntry("ollama-cloud",   "Ollama Cloud",             "Ollama Cloud (Cloud-hosted open models, ollama.com)"),
     ProviderEntry("arcee",          "Arcee AI",                 "Arcee AI (Trinity models, direct API)"),
+    ProviderEntry("featherless",    "Featherless",              "Featherless (Open-source models via HF repo IDs, direct API)"),
     ProviderEntry("gmi",            "GMI Cloud",                "GMI Cloud (Multi-model direct API)"),
     ProviderEntry("kilocode",       "Kilo Code",                "Kilo Code (Kilo Gateway API)"),
     ProviderEntry("opencode-zen",   "OpenCode Zen",             "OpenCode Zen (Curated models, pay-as-you-go)"),
@@ -1198,6 +1207,8 @@ def group_providers(slugs):
     "stepfun-coding-plan": "stepfun",
     "arcee-ai": "arcee",
     "arceeai": "arcee",
+    "featherless-ai": "featherless",
+    "featherlessai": "featherless",
     "gmi-cloud": "gmi",
     "gmicloud": "gmi",
     "minimax-china": "minimax-cn",

hermes_cli/providers.py

@@ -190,6 +190,11 @@ class HermesOverlay:
         base_url_override="https://api.arcee.ai/api/v1",
         base_url_env_var="ARCEE_BASE_URL",
     ),
+    "featherless": HermesOverlay(
+        transport="openai_chat",
+        base_url_override="https://api.featherless.ai/v1",
+        base_url_env_var="FEATHERLESS_BASE_URL",
+    ),
     "gmi": HermesOverlay(
         transport="openai_chat",
         extra_env_vars=("GMI_API_KEY",),
@@ -344,6 +349,10 @@ class ProviderDef:
     "arcee-ai": "arcee",
     "arceeai": "arcee",
 
+    # featherless
+    "featherless-ai": "featherless",
+    "featherlessai": "featherless",
+
     # gmi
     "gmi-cloud": "gmi",
     "gmicloud": "gmi",

hermes_cli/setup.py

@@ -98,6 +98,7 @@ def _supports_same_provider_pool_setup(provider: str) -> bool:
     "kimi-coding-cn": ["kimi-k2.6", "kimi-k2.5", "kimi-k2-thinking", "kimi-k2-turbo-preview"],
     "stepfun": ["step-3.5-flash", "step-3.5-flash-2603"],
     "arcee": ["trinity-large-thinking", "trinity-large-preview", "trinity-mini"],
+    "featherless": ["zai-org/GLM-5.2", "zai-org/GLM-4.7", "moonshotai/Kimi-K2-Thinking", "deepseek-ai/DeepSeek-V3.2", "Qwen/Qwen3.5-397B-A17B", "meta-llama/Llama-3.1-8B-Instruct"],
     "minimax": ["MiniMax-M2.7", "MiniMax-M2.5", "MiniMax-M2.1", "MiniMax-M2"],
     "minimax-cn": ["MiniMax-M2.7", "MiniMax-M2.5", "MiniMax-M2.1", "MiniMax-M2"],
     "kilocode": ["anthropic/claude-opus-4.6", "anthropic/claude-sonnet-4.6", "openai/gpt-5.4", "google/gemini-3-pro-preview", "google/gemini-3-flash-preview"],

plugins/model-providers/featherless/__init__.py

@@ -0,0 +1,13 @@
+"""Featherless provider profile."""
+
+from providers import register_provider
+from providers.base import ProviderProfile
+
+featherless = ProviderProfile(
+    name="featherless",
+    aliases=("featherless-ai", "featherlessai"),
+    env_vars=("FEATHERLESS_API_KEY",),
+    base_url="https://api.featherless.ai/v1",
+)
+
+register_provider(featherless)

plugins/model-providers/featherless/plugin.yaml

@@ -0,0 +1,5 @@
+name: featherless-provider
+kind: model-provider
+version: 1.0.0
+description: Featherless
+author: Nous Research