NousResearch · kshitijk4poor · Apr 8, 2026
diff --git a/.env.example b/.env.example
@@ -24,6 +24,15 @@
 # Optional base URL override (default: Google's OpenAI-compatible endpoint)
 # GEMINI_BASE_URL=https://generativelanguage.googleapis.com/v1beta/openai
 
+# =============================================================================
+# LLM PROVIDER (Fireworks AI)
+# =============================================================================
+# Fast serverless inference. OpenAI-compatible endpoint.
+# Get your key at: https://fireworks.ai/account/api-keys
+# FIREWORKS_API_KEY=your_fireworks_key_here
+# Optional base URL override (default: https://api.fireworks.ai/inference/v1)
+# FIREWORKS_BASE_URL=https://api.fireworks.ai/inference/v1
+
 # =============================================================================
 # LLM PROVIDER (z.ai / GLM)
 # =============================================================================

diff --git a/agent/auxiliary_client.py b/agent/auxiliary_client.py
@@ -106,6 +106,7 @@ def _normalize_aux_provider(provider: Optional[str], *, for_vision: bool = False
     "opencode-zen": "gemini-3-flash",
     "opencode-go": "glm-5",
     "kilocode": "google/gemini-3-flash-preview",
+    "fireworks": "accounts/fireworks/models/llama3.1-8b",
 }
 
 # OpenRouter app attribution headers

diff --git a/agent/model_metadata.py b/agent/model_metadata.py
@@ -24,13 +24,14 @@
 # are preserved so the full model name reaches cache lookups and server queries.
 _PROVIDER_PREFIXES: frozenset[str] = frozenset({
     "openrouter", "nous", "openai-codex", "copilot", "copilot-acp",
-    "gemini", "zai", "kimi-coding", "minimax", "minimax-cn", "anthropic", "deepseek",
+    "gemini", "fireworks", "zai", "kimi-coding", "minimax", "minimax-cn", "anthropic", "deepseek",
     "opencode-zen", "opencode-go", "ai-gateway", "kilocode", "alibaba",
     "custom", "local",
     # Common aliases
     "google", "google-gemini", "google-ai-studio",
     "glm", "z-ai", "z.ai", "zhipu", "github", "github-copilot",
     "github-models", "kimi", "moonshot", "claude", "deep-seek",
+    "fw", "fireworks-ai",
     "opencode", "zen", "go", "vercel", "kilo", "dashscope", "aliyun", "qwen",
 })
 

diff --git a/cli-config.yaml.example b/cli-config.yaml.example
@@ -24,6 +24,7 @@ model:
   #   "minimax"      - MiniMax global (requires: MINIMAX_API_KEY)
   #   "minimax-cn"   - MiniMax China (requires: MINIMAX_CN_API_KEY)
   #   "huggingface"  - Hugging Face Inference (requires: HF_TOKEN)
+  #   "fireworks"    - Fireworks AI (requires: FIREWORKS_API_KEY — https://fireworks.ai/account/api-keys)
   #   "kilocode"     - KiloCode gateway (requires: KILOCODE_API_KEY)
   #   "ai-gateway"   - Vercel AI Gateway (requires: AI_GATEWAY_API_KEY)
   #
@@ -317,6 +318,7 @@ compression:
 #   "openrouter" - Force OpenRouter (requires OPENROUTER_API_KEY)
 #   "nous"       - Force Nous Portal (requires: hermes login)
 #   "gemini"      - Force Google AI Studio direct (requires: GOOGLE_API_KEY or GEMINI_API_KEY)
+#   "fireworks"   - Fireworks AI (requires: FIREWORKS_API_KEY)
 #   "codex"       - Force Codex OAuth (requires: hermes model → Codex).
 #                  Uses gpt-5.3-codex which supports vision.
 #   "main"       - Use your custom endpoint (OPENAI_BASE_URL + OPENAI_API_KEY).

diff --git a/hermes_cli/auth.py b/hermes_cli/auth.py
@@ -70,6 +70,7 @@
 DEFAULT_GITHUB_MODELS_BASE_URL = "https://api.githubcopilot.com"
 DEFAULT_COPILOT_ACP_BASE_URL = "acp://copilot"
 DEFAULT_GEMINI_BASE_URL = "https://generativelanguage.googleapis.com/v1beta/openai"
+DEFAULT_FIREWORKS_BASE_URL = "https://api.fireworks.ai/inference/v1"
 CODEX_OAUTH_CLIENT_ID = "app_EMoamEEZ73f0CkXaXp7hrann"
 CODEX_OAUTH_TOKEN_URL = "https://auth.openai.com/oauth/token"
 CODEX_ACCESS_TOKEN_REFRESH_SKEW_SECONDS = 120
@@ -233,6 +234,14 @@ class ProviderConfig:
         api_key_env_vars=("HF_TOKEN",),
         base_url_env_var="HF_BASE_URL",
     ),
+    "fireworks": ProviderConfig(
+        id="fireworks",
+        name="Fireworks AI",
+        auth_type="api_key",
+        inference_base_url=DEFAULT_FIREWORKS_BASE_URL,
+        api_key_env_vars=("FIREWORKS_API_KEY",),
+        base_url_env_var="FIREWORKS_BASE_URL",
+    ),
 }
 
 
@@ -820,6 +829,7 @@ def resolve_provider(
         "hf": "huggingface", "hugging-face": "huggingface", "huggingface-hub": "huggingface",
         "go": "opencode-go", "opencode-go-sub": "opencode-go",
         "kilo": "kilocode", "kilo-code": "kilocode", "kilo-gateway": "kilocode",
+        "fireworks-ai": "fireworks", "fw": "fireworks",
         # Local server aliases — route through the generic custom provider
         "lmstudio": "custom", "lm-studio": "custom", "lm_studio": "custom",
         "ollama": "custom", "vllm": "custom", "llamacpp": "custom",

diff --git a/hermes_cli/config.py b/hermes_cli/config.py
@@ -771,6 +771,22 @@ def ensure_hermes_home():
         "category": "provider",
         "advanced": True,
     },
+    "FIREWORKS_API_KEY": {
+        "description": "Fireworks AI API key (fast serverless inference)",
+        "prompt": "Fireworks AI API key",
+        "url": "https://fireworks.ai/account/api-keys",
+        "password": True,
+        "category": "provider",
+        "advanced": True,
+    },
+    "FIREWORKS_BASE_URL": {
+        "description": "Fireworks AI base URL override (default: https://api.fireworks.ai/inference/v1)",
+        "prompt": "Fireworks base URL (leave empty for default)",
+        "url": None,
+        "password": False,
+        "category": "provider",
+        "advanced": True,
+    },
 
     # ── Tool API keys ──
     "EXA_API_KEY": {

diff --git a/hermes_cli/main.py b/hermes_cli/main.py
@@ -932,6 +932,7 @@ def select_provider_and_model(args=None):
         "kilocode": "Kilo Code",
         "alibaba": "Alibaba Cloud (DashScope)",
         "huggingface": "Hugging Face",
+        "fireworks": "Fireworks AI",
         "custom": "Custom endpoint",
     }
     active_label = provider_labels.get(active, active) if active else "none"
@@ -963,6 +964,7 @@ def select_provider_and_model(args=None):
         ("opencode-go", "OpenCode Go (open models, $10/month subscription)"),
         ("ai-gateway", "AI Gateway (Vercel — 200+ models, pay-per-use)"),
         ("alibaba", "Alibaba Cloud / DashScope Coding (Qwen + multi-provider)"),
+        ("fireworks", "Fireworks AI (fast serverless inference — OpenAI-compatible)"),
     ]
 
     # Add user-defined custom providers from config.yaml
@@ -1057,7 +1059,7 @@ def select_provider_and_model(args=None):
         _model_flow_anthropic(config, current_model)
     elif selected_provider == "kimi-coding":
         _model_flow_kimi(config, current_model)
-    elif selected_provider in ("gemini", "zai", "minimax", "minimax-cn", "kilocode", "opencode-zen", "opencode-go", "ai-gateway", "alibaba", "huggingface"):
+    elif selected_provider in ("gemini", "zai", "minimax", "minimax-cn", "kilocode", "opencode-zen", "opencode-go", "ai-gateway", "alibaba", "huggingface", "fireworks"):
         _model_flow_api_key_provider(config, selected_provider, current_model)
 
 
@@ -4240,7 +4242,7 @@ def main():
     )
     chat_parser.add_argument(
         "--provider",
-        choices=["auto", "openrouter", "nous", "openai-codex", "copilot-acp", "copilot", "anthropic", "gemini", "huggingface", "zai", "kimi-coding", "minimax", "minimax-cn", "kilocode"],
+        choices=["auto", "openrouter", "nous", "openai-codex", "copilot-acp", "copilot", "anthropic", "gemini", "fireworks", "huggingface", "zai", "kimi-coding", "minimax", "minimax-cn", "kilocode"],
         default=None,
         help="Inference provider (default: auto)"
     )

diff --git a/hermes_cli/model_normalize.py b/hermes_cli/model_normalize.py
@@ -85,6 +85,7 @@
     "minimax-cn",
     "alibaba",
     "huggingface",
+    "fireworks",
     "openai-codex",
     "custom",
 })

diff --git a/hermes_cli/models.py b/hermes_cli/models.py
@@ -480,6 +480,7 @@ def check_nous_free_tier() -> bool:
     "kilocode": "Kilo Code",
     "alibaba": "Alibaba Cloud (DashScope)",
     "huggingface": "Hugging Face",
+    "fireworks": "Fireworks AI",
     "custom": "Custom endpoint",
 }
 
@@ -521,6 +522,8 @@ def check_nous_free_tier() -> bool:
     "hf": "huggingface",
     "hugging-face": "huggingface",
     "huggingface-hub": "huggingface",
+    "fireworks-ai": "fireworks",
+    "fw": "fireworks",
 }
 
 
@@ -761,7 +764,7 @@ def list_available_providers() -> list[dict[str, str]]:
     # Canonical providers in display order
     _PROVIDER_ORDER = [
         "openrouter", "nous", "openai-codex", "copilot", "copilot-acp",
-        "gemini", "huggingface",
+        "gemini", "fireworks", "huggingface",
         "zai", "kimi-coding", "minimax", "minimax-cn", "kilocode", "anthropic", "alibaba",
         "opencode-zen", "opencode-go",
         "ai-gateway", "deepseek", "custom",

diff --git a/hermes_cli/providers.py b/hermes_cli/providers.py
@@ -121,6 +121,10 @@ class HermesOverlay:
         is_aggregator=True,
         base_url_env_var="HF_BASE_URL",
     ),
+    "fireworks": HermesOverlay(
+        transport="openai_chat",
+        base_url_env_var="FIREWORKS_BASE_URL",
+    ),
 }
 
 
@@ -200,6 +204,10 @@ def is_user_defined(self) -> bool:
     # deepseek
     "deep-seek": "deepseek",
 
+    # fireworks
+    "fireworks-ai": "fireworks",
+    "fw": "fireworks",
+
     # alibaba
     "dashscope": "alibaba",
     "aliyun": "alibaba",
@@ -232,6 +240,7 @@ def is_user_defined(self) -> bool:
     "openai-codex": "OpenAI Codex",
     "copilot-acp": "GitHub Copilot ACP",
     "local": "Local endpoint",
+    "fireworks": "Fireworks AI",
 }
 
 
@@ -367,6 +376,7 @@ def get_label(provider_id: str) -> str:
     "huggingface": "Hugging Face",
     "local": "Local endpoint",
     "custom": "Custom endpoint",
+    "fireworks": "Fireworks AI",
     # Legacy Hermes IDs (point to same providers)
     "ai-gateway": "Vercel AI Gateway",
     "kilocode": "Kilo Gateway",