Skip to content

Use Claude Code, Codex, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from your terminal, app, IDE, or phone, and now from the browser with native browser sessions (multi-harness + multi-model) like OpenClaw (voice supported + ToS friendly)

Python

Add Lightning AI provider (#1732) (#1740)

Tanishq-1613cdaaa6bd6
diff --git a/.env.example b/.env.example
index 3322090d..b01e20b2 100644
--- a/.env.example
+++ b/.env.example
@@ -83,6 +83,11 @@ NARAROUTE_BASE_URL="https://router.bynara.id/v1"
# LLM7_API_KEY=<your-token>
+# Lightning AI Model APIs (OpenAI-compatible Chat Completions at lightning.ai/api/v1)
+# LIGHTNING_API_KEY=<your-token>
+# LIGHTNING_BASE_URL="https://lightning.ai/api/v1"
+
+
# Fireworks AI Config (OpenAI-compatible Chat Completions at api.fireworks.ai/inference/v1)
# FIREWORKS_API_KEY=<your-token>
@@ -196,7 +201,7 @@ OLLAMA_BASE_URL="http://localhost:11434"
# Default model and optional Claude tier overrides
# Format: provider_type/model/name
-# Valid providers: "nvidia_nim" | "openai" | "azure_openai" | "open_router" | "gemini" | "vertex" | "deepseek" | "mistral" | "mistral_codestral" | "opencode_zen" | "opencode_go" | "vercel" | "bedrock" | "huggingface" | "cohere" | "wafer" | "kimi" | "kimi_code" | "minimax" | "cerebras" | "groq" | "cline_pass" | "xai" | "qwencloud" | "qwencloud_coding" | "together" | "deepinfra" | "siliconflow" | "nebius" | "chutes" | "featherless" | "agnes" | "zenmux" | "wandb" | "sambanova" | "kilo" | "fireworks" | "novita" | "cloudflare" | "zai" | "ollama_cloud" | "lmstudio" | "llamacpp" | "ollama" | "tokenrouter" | "nararoute" | "poolside" | "llm7"
+# Valid providers: "nvidia_nim" | "openai" | "azure_openai" | "open_router" | "gemini" | "vertex" | "deepseek" | "mistral" | "mistral_codestral" | "opencode_zen" | "opencode_go" | "vercel" | "bedrock" | "huggingface" | "cohere" | "wafer" | "kimi" | "kimi_code" | "minimax" | "cerebras" | "groq" | "cline_pass" | "xai" | "qwencloud" | "qwencloud_coding" | "together" | "deepinfra" | "siliconflow" | "nebius" | "chutes" | "featherless" | "agnes" | "zenmux" | "wandb" | "sambanova" | "kilo" | "fireworks" | "novita" | "cloudflare" | "zai" | "ollama_cloud" | "lmstudio" | "llamacpp" | "ollama" | "tokenrouter" | "nararoute" | "poolside" | "llm7" | "lightning"
# MODEL_FABLE=<provider/model-id>
# MODEL_OPUS=<provider/model-id>
# MODEL_SONNET=<provider/model-id>
@@ -238,6 +243,7 @@ MODEL="nvidia_nim/nvidia/nemotron-3-super-120b-a12b"
# FCC_SMOKE_MODEL_NARAROUTE=<provider/model-id>
# FCC_SMOKE_MODEL_POOLSIDE=<provider/model-id>
# FCC_SMOKE_MODEL_LLM7=<provider/model-id>
+# FCC_SMOKE_MODEL_LIGHTNING=<provider/model-id>
# FCC_SMOKE_MODEL_FIREWORKS=<provider/model-id>
# FCC_SMOKE_MODEL_NOVITA=<provider/model-id>
# FCC_SMOKE_MODEL_CLOUDFLARE=<provider/model-id>
@@ -316,6 +322,7 @@ REASONING_HAIKU=inherit
# NARAROUTE_PROXY=<http-or-socks-url>
# POOLSIDE_PROXY=<http-or-socks-url>
# LLM7_PROXY=<http-or-socks-url>
+# LIGHTNING_PROXY=<http-or-socks-url>
# FIREWORKS_PROXY=<http-or-socks-url>
# NOVITA_PROXY=<http-or-socks-url>
# CLOUDFLARE_PROXY=<http-or-socks-url>
diff --git a/README.md b/README.md
index 335b8645..d615c9cd 100644
--- a/README.md
+++ b/README.md
@@ -25,7 +25,7 @@
## What You Get
-- **50 ToS-friendly providers. 1.3B+ free tokens every month.** Use free, paid, subscription, and local models from one searchable UI without putting your account at risk. FCC follows provider terms and removes integrations if they stop being allowed.
+- **52 ToS-friendly providers. 1.3B+ free tokens every month.** Use free, paid, subscription, and local models from one searchable UI without putting your account at risk. FCC follows provider terms and removes integrations if they stop being allowed.
- **10 coding agents. One model catalog.** Run [Claude Code](https://code.claude.com/docs/en/overview), [Codex](https://github.com/openai/codex), [Pi](https://github.com/earendil-works/pi), [OpenCode](https://github.com/anomalyco/opencode), [Cline](https://github.com/cline/cline), [Hermes](https://github.com/NousResearch/hermes-agent), [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness), [Grok Build](https://github.com/xai-org/grok-build), [Muse Code](https://research.meta.ai/blog/introducing-muse-code-and-muse-spark-1-2/), or [Aider](https://aider.chat/) with your FCC models.
- **Keep coding through provider outages.** After retries are exhausted, FCC automatically tries your next configured model without making you restart the turn—across every client.
- **Up to 90% fewer terminal-output tokens.** Optional [RTK](https://github.com/rtk-ai/rtk) filters common command output, while five FCC optimizations handle quota probes, command-prefix detection, titles, suggestions, and filepaths without calling a provider.
@@ -238,6 +238,8 @@ from more than one provider before succeeding.
| [NaraRoute](https://router.bynara.id/) | `NARAROUTE_API_KEY` | `nararoute/kimi-k3-free` |
| [Poolside AI](https://platform.poolside.ai/) | `POOLSIDE_API_KEY` | `poolside/poolside/laguna-s-2.1` |
| [LLM7.io](https://dash.llm7.io/) | `LLM7_API_KEY` | `llm7/default` |
+| [Scaleway](https://console.scaleway.com/iam/api-keys) | `SCW_SECRET_KEY` | `scaleway/deepseek/deepseek-v4-flash` |
+| [Lightning AI](https://lightning.ai/) | `LIGHTNING_API_KEY` | `lightning/lightning-ai/Qwen3.8-27B` |
| [Ollama Cloud](https://ollama.com/settings/keys) | `OLLAMA_API_KEY` | `ollama_cloud/qwen3-coder:480b` |
| [LM Studio](https://lmstudio.ai/) | `LM_STUDIO_BASE_URL` | `lmstudio/<model-id>` |
| [llama.cpp](https://github.com/ggml-org/llama.cpp) | `LLAMACPP_BASE_URL` | `llamacpp/<model-id>` |
diff --git a/pyproject.toml b/pyproject.toml
index 91162088..f17d97ae 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
[project]
name = "free-claude-code"
-version = "6.2.37"
+version = "6.2.38"
description = "Local proxy connecting coding agents to OpenAI-compatible AI providers"
readme = "README.md"
requires-python = ">=3.14.0"
diff --git a/smoke/lib/config.py b/smoke/lib/config.py
index 45ac4001..11c5aa69 100644
--- a/smoke/lib/config.py
+++ b/smoke/lib/config.py
@@ -93,6 +93,7 @@ PROVIDER_SMOKE_DEFAULT_MODELS: dict[str, str] = {
"nararoute": "nararoute/kimi-k3-free",
"poolside": "poolside/poolside/laguna-s-2.1",
"llm7": "llm7/default",
+ "lightning": "lightning/lightning-ai/Qwen3.8-27B",
"agnes": "agnes/agnes-2.0-flash",
"zenmux": "zenmux/deepseek/deepseek-v4-flash-free",
"wandb": "wandb/openai/gpt-oss-20b",
diff --git a/src/free_claude_code/api/admin_static/providers/SOURCES.md b/src/free_claude_code/api/admin_static/providers/SOURCES.md
index 1d911881..67a13d18 100644
--- a/src/free_claude_code/api/admin_static/providers/SOURCES.md
+++ b/src/free_claude_code/api/admin_static/providers/SOURCES.md
@@ -25,6 +25,7 @@ The LLaMA.cpp icon comes from the upstream llama.cpp repository; its MIT license
| `huggingface-color.svg` | https://unpkg.com/@lobehub/[email protected]/icons/huggingface-color.svg |
| `kilocode.svg` | https://unpkg.com/@lobehub/[email protected]/icons/kilocode.svg |
| `kimi-color.svg` | https://unpkg.com/@lobehub/[email protected]/icons/kimi-color.svg |
+| `lightning.png` | https://lightning.ai/favicon/favicon-32x32.png |
| `llamacpp.svg` | https://raw.githubusercontent.com/ggml-org/llama.cpp/master/media/llama1-icon-transparent.svg |
| `llm7.png` | https://llm7.io/favicons/apple-touch-icon.png |
| `lmstudio.svg` | https://unpkg.com/@lobehub/[email protected]/icons/lmstudio.svg |
diff --git a/src/free_claude_code/api/admin_static/providers/lightning.png b/src/free_claude_code/api/admin_static/providers/lightning.png
new file mode 100644
index 00000000..b2e47f08
Binary files /dev/null and b/src/free_claude_code/api/admin_static/providers/lightning.png differ
diff --git a/src/free_claude_code/config/admin/provider_manifest.py b/src/free_claude_code/config/admin/provider_manifest.py
index 77b273ac..f065094a 100644
--- a/src/free_claude_code/config/admin/provider_manifest.py
+++ b/src/free_claude_code/config/admin/provider_manifest.py
@@ -270,6 +270,19 @@ _PROVIDER_FIELD_OVERRIDES: dict[str, ProviderFieldOverride] = {
"Defaults to https://router.bynara.id/v1."
),
},
+ "LIGHTNING_API_KEY": {
+ "label": "Lightning AI API Key",
+ "description": (
+ "Lightning AI Model APIs key for the OpenAI-compatible endpoint at "
+ "lightning.ai/api/v1. Create one on lightning.ai under Model APIs."
+ ),
+ },
+ "LIGHTNING_BASE_URL": {
+ "description": (
+ "Lightning AI OpenAI-compatible Chat Completions base URL. "
+ "Defaults to https://lightning.ai/api/v1."
+ ),
+ },
"AGNES_API_KEY": {
"label": "Agnes AI API Key",
"description": (
diff --git a/src/free_claude_code/config/provider_catalog.py b/src/free_claude_code/config/provider_catalog.py
index c12a6da0..393bb58e 100644
--- a/src/free_claude_code/config/provider_catalog.py
+++ b/src/free_claude_code/config/provider_catalog.py
@@ -82,6 +82,8 @@ NARAROUTE_DEFAULT_BASE = "https://router.bynara.id/v1"
POOLSIDE_DEFAULT_BASE = "https://inference.poolside.ai/v1"
# LLM7.io OpenAI-compatible Chat Completions API.
LLM7_DEFAULT_BASE = "https://api.llm7.io/v1"
+# Lightning AI Model APIs OpenAI-compatible Chat Completions gateway.
+LIGHTNING_DEFAULT_BASE = "https://lightning.ai/api/v1"
# Agnes AI OpenAI-compatible Chat Completions API.
AGNES_DEFAULT_BASE = "https://apihub.agnes-ai.com/v1"
# ZenMux OpenAI-compatible Chat Completions gateway.
@@ -652,6 +654,18 @@ PROVIDER_CATALOG: dict[str, ProviderDescriptor] = {
default_base_url=SCALEWAY_DEFAULT_BASE,
proxy_attr="scw_proxy",
),
+ "lightning": ProviderDescriptor(
+ provider_id="lightning",
+ display_name="Lightning AI",
+ website_url="https://lightning.ai/",
+ logo_filename="lightning.png",
+ credential_env="LIGHTNING_API_KEY",
+ credential_url="https://lightning.ai/lightning-ai/model-apis/models",
+ credential_attr="lightning_api_key",
+ default_base_url=LIGHTNING_DEFAULT_BASE,
+ base_url_attr="lightning_base_url",
+ proxy_attr="lightning_proxy",
+ ),
"ollama_cloud": ProviderDescriptor(
provider_id="ollama_cloud",
display_name="Ollama Cloud",
diff --git a/src/free_claude_code/config/settings.py b/src/free_claude_code/config/settings.py
index 3785393e..d7d176f7 100644
--- a/src/free_claude_code/config/settings.py
+++ b/src/free_claude_code/config/settings.py
@@ -17,6 +17,7 @@ from .model_refs import parse_model_fallbacks
from .nim import NimSettings
from .provider_catalog import (
BEDROCK_DEFAULT_BASE,
+ LIGHTNING_DEFAULT_BASE,
NARAROUTE_DEFAULT_BASE,
SUPPORTED_PROVIDER_IDS,
TOKENROUTER_DEFAULT_BASE,
@@ -190,6 +191,15 @@ class Settings(BaseModel):
default=None, validation_alias="LLM7_API_KEY"
)
+ # ==================== Lightning AI (OpenAI-compatible) ====================
+ lightning_api_key: OptionalNonEmptyString = Field(
+ default=None, validation_alias="LIGHTNING_API_KEY"
+ )
+ lightning_base_url: NonEmptyString = Field(
+ default=LIGHTNING_DEFAULT_BASE,
+ validation_alias="LIGHTNING_BASE_URL",
+ )
+
# ==================== Fireworks AI Config ====================
fireworks_api_key: OptionalNonEmptyString = Field(
default=None, validation_alias="FIREWORKS_API_KEY"
@@ -484,6 +494,9 @@ class Settings(BaseModel):
llm7_proxy: OptionalNonEmptyString = Field(
default=None, validation_alias="LLM7_PROXY"
)
+ lightning_proxy: OptionalNonEmptyString = Field(
+ default=None, validation_alias="LIGHTNING_PROXY"
+ )
fireworks_proxy: OptionalNonEmptyString = Field(
default=None, validation_alias="FIREWORKS_PROXY"
)
diff --git a/src/free_claude_code/providers/openai_chat/profiles.py b/src/free_claude_code/providers/openai_chat/profiles.py
index b791eb26..f4f5cf84 100644
--- a/src/free_claude_code/providers/openai_chat/profiles.py
+++ b/src/free_claude_code/providers/openai_chat/profiles.py
@@ -694,6 +694,23 @@ OPENAI_CHAT_PROFILES: dict[str, OpenAIChatProfile] = {
context_window_tokens_path=("context_window", "tokens"),
),
),
+ "lightning": OpenAIChatProfile(
+ _policy(
+ "LIGHTNING",
+ ReasoningReplayMode.DISABLED,
+ default_max_tokens=ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS,
+ ),
+ NamedEffortReasoning(
+ _LOW_MEDIUM_HIGH,
+ disabled_value="none",
+ enabled_value="medium",
+ ),
+ model_listing=OpenAIModelListing(
+ context_window_tokens_path=("context_length",),
+ max_output_tokens_path=("max_tokens",),
+ input_modalities_path=("architecture", "input_modalities"),
+ ),
+ ),
"ollama_cloud": OpenAIChatProfile(
_policy(
"OLLAMA_CLOUD",
diff --git a/tests/contracts/test_provider_catalog_order.py b/tests/contracts/test_provider_catalog_order.py
index 2afcc3b1..7d7cde27 100644
--- a/tests/contracts/test_provider_catalog_order.py
+++ b/tests/contracts/test_provider_catalog_order.py
@@ -53,6 +53,7 @@ _EXPECTED_PROVIDER_ORDER: tuple[str, ...] = (
"poolside",
"llm7",
"scaleway",
+ "lightning",
"ollama_cloud",
"lmstudio",
"llamacpp",
diff --git a/tests/contracts/test_smoke_config.py b/tests/contracts/test_smoke_config.py
index 5a6c2f27..21bc2caf 100644
--- a/tests/contracts/test_smoke_config.py
+++ b/tests/contracts/test_smoke_config.py
@@ -68,6 +68,7 @@ def _settings(**overrides):
"ollama_api_key": "",
"poolside_api_key": "",
"llm7_api_key": "",
+ "lightning_api_key": "",
"fireworks_api_key": "",
"novita_api_key": "",
"cloudflare_api_token": "",
@@ -233,6 +234,56 @@ def test_llm7_is_not_enabled_without_explicit_credential(monkeypatch) -> None:
assert config.provider_smoke_models() == []
+def test_lightning_provider_configuration_uses_documented_default_model(
+ monkeypatch,
+) -> None:
+ monkeypatch.delenv("FCC_SMOKE_MODEL_LIGHTNING", raising=False)
+ config = _smoke_config(
+ settings=_settings(
+ model="ollama/llama3.1",
+ ollama_base_url="",
+ lightning_api_key="lightning-key",
+ )
+ )
+
+ assert config.has_provider_configuration("lightning")
+ models = config.provider_smoke_models()
+ assert [model.provider for model in models] == ["lightning"]
+ assert models[0].full_model == "lightning/lightning-ai/Qwen3.8-27B"
+ assert models[0].source == "provider_default"
+
+
+def test_lightning_smoke_override_accepts_model_with_or_without_prefix(
+ monkeypatch,
+) -> None:
+ settings = _settings(
+ model="ollama/llama3.1",
+ ollama_base_url="",
+ lightning_api_key="lightning-key",
+ )
+ for override in (
+ "lightning-ai/deepseek-v4-pro",
+ "lightning/lightning-ai/deepseek-v4-pro",
+ ):
+ monkeypatch.setenv("FCC_SMOKE_MODEL_LIGHTNING", override)
+ models = _smoke_config(settings=settings).provider_smoke_models()
+
+ assert [model.provider for model in models] == ["lightning"]
+ assert models[0].full_model == "lightning/lightning-ai/deepseek-v4-pro"
+ assert models[0].source == "FCC_SMOKE_MODEL_LIGHTNING"
+
+
+def test_lightning_is_not_enabled_without_explicit_credential(monkeypatch) -> None:
+ monkeypatch.delenv("FCC_SMOKE_MODEL_LIGHTNING", raising=False)
+ config = _smoke_config(
+ provider_matrix=frozenset({"lightning"}),
+ settings=_settings(ollama_base_url="", lightning_api_key=""),
+ )
+
+ assert not config.has_provider_configuration("lightning")
+ assert config.provider_smoke_models() == []
+
+
def test_xai_provider_smoke_uses_current_grok_model(monkeypatch) -> None:
monkeypatch.delenv("FCC_SMOKE_MODEL_XAI", raising=False)
config = _smoke_config(
diff --git a/tests/providers/test_credential_validation.py b/tests/providers/test_credential_validation.py
index 651836bc..06e2f8e4 100644
--- a/tests/providers/test_credential_validation.py
+++ b/tests/providers/test_credential_validation.py
@@ -197,11 +197,11 @@ async def test_unsupported_shared_credentials_never_send_http(monkeypatch):
supported = {PROVIDER_CATALOG[row[0]].credential_env for row in CASES}
all_keys = {d.credential_env for d in PROVIDER_CATALOG.values() if d.credential_env}
keys = tuple(all_keys - supported)
- assert len(keys) == 20
+ assert len(keys) == 21
result = await validation.check_credentials(
Settings.model_construct(), (*keys, "OPENCODE_API_KEY", "MODEL")
)
- assert len(result) == 20
+ assert len(result) == 21
assert all(
check.status == validation.CredentialStatus.UNVERIFIED for check in result
)
diff --git a/tests/providers/test_lightning.py b/tests/providers/test_lightning.py
new file mode 100644
index 00000000..19451cc9
--- /dev/null
+++ b/tests/providers/test_lightning.py
@@ -0,0 +1,238 @@
+"""Tests for the Lightning AI Model APIs OpenAI-chat provider profile."""
+
+from unittest.mock import AsyncMock
+
+import httpx2
+import pytest
+from openai import AsyncOpenAI
+
+from free_claude_code.application.model_metadata import ProviderModelInfo
+from free_claude_code.config.constants import ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS
+from free_claude_code.config.provider_catalog import LIGHTNING_DEFAULT_BASE
+from free_claude_code.core.anthropic.models import MessagesRequest
+from free_claude_code.core.json_types import JsonObject, JsonValue
+from free_claude_code.core.model_capabilities import ModelInputModality
+from free_claude_code.core.reasoning import ReasoningEffort, ReasoningPolicy
+from free_claude_code.providers.model_listing import ModelListResponseError
+from free_claude_code.providers.openai_chat import OpenAIChatProvider
+from free_claude_code.providers.openai_chat.profiles import OPENAI_CHAT_PROFILES
+from tests.providers.support import (
+ REASONING_DEFAULT,
+ REASONING_OFF,
+ REASONING_ON,
+ immediate_admission,
+ make_provider_config,
+ profiled_provider,
+)
+
+_MODEL = "lightning-ai/Qwen3.8-27B"
+
+
+def lightning_provider() -> OpenAIChatProvider:
+ return profiled_provider(
+ "lightning",
+ make_provider_config(
+ api_key="test-lightning-key",
+ base_url=LIGHTNING_DEFAULT_BASE,
+ rate_limit=10,
+ rate_window=60,
+ ),
+ admission=immediate_admission(provider_name="lightning", max_attempts=1),
+ )
+
+
+def _request(**overrides: JsonValue) -> MessagesRequest:
+ payload: JsonObject = {
+ "model": _MODEL,
+ "messages": [{"role": "user", "content": "Inspect the file."}],
+ "tools": [
+ {
+ "name": "read_file",
+ "description": "Read a file",
+ "input_schema": {
+ "type": "object",
+ "properties": {"path": {"type": "string"}},
+ "required": ["path"],
+ },
+ }
+ ],
+ }
+ payload.update(overrides)
+ return MessagesRequest.model_validate(payload)
+
+
+def _catalog_model(
+ model_id: str = _MODEL,
+ *,
+ context_length: int | None = None,
+ max_tokens: int | None = None,
+ input_modalities: object = ("text",),
+) -> dict[str, object]:
+ model: dict[str, object] = {"id": model_id}
+ if input_modalities is not None:
+ model["architecture"] = {"input_modalities": input_modalities}
+ if context_length is not None:
+ model["context_length"] = context_length
+ if max_tokens is not None:
+ model["max_tokens"] = max_tokens
+ return model
+
+
+def test_constructs_standard_openai_chat_provider(
+ lightning_provider: OpenAIChatProvider,
+) -> None:
+ assert isinstance(lightning_provider, OpenAIChatProvider)
+ assert lightning_provider._provider_name == "LIGHTNING"
+ assert lightning_provider._api_key == "test-lightning-key"
+ assert lightning_provider._base_url == LIGHTNING_DEFAULT_BASE
+
+
+ ("reasoning", "expected_reasoning_effort"),
+ [
+ (REASONING_DEFAULT, None),
+ (REASONING_ON, "medium"),
+ (REASONING_OFF, "none"),
+ ],
+)
+def test_encodes_documented_reasoning_effort_and_standard_request_fields(
+ lightning_provider: OpenAIChatProvider,
+ reasoning: ReasoningPolicy,
+ expected_reasoning_effort: str | None,
+) -> None:
+ body = lightning_provider._chat._build_request_body(_request(), reasoning=reasoning)
+
+ assert body["max_tokens"] == ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS
+ assert body["model"] == _MODEL
+ assert body["tools"][0]["function"]["name"] == "read_file"
+ if expected_reasoning_effort is None:
+ assert "reasoning_effort" not in body
+ else:
+ assert body["reasoning_effort"] == expected_reasoning_effort
+ assert "reasoning" not in body
+ assert "thinking" not in body
+ assert "extra_body" not in body
+
+
+ ("policy", "expected"),
+ [
+ (ReasoningPolicy.on(effort=ReasoningEffort.MINIMAL), "low"),
+ (ReasoningPolicy.on(effort=ReasoningEffort.LOW), "low"),
+ (ReasoningPolicy.on(effort=ReasoningEffort.MEDIUM), "medium"),
+ (ReasoningPolicy.on(effort=ReasoningEffort.HIGH), "high"),
+ (ReasoningPolicy.on(effort=ReasoningEffort.XHIGH), "high"),
+ (ReasoningPolicy.on(effort=ReasoningEffort.MAX), "high"),
+ ],
+)
+def test_reasoning_uses_lightning_documented_effort_vocabulary(
+ lightning_provider: OpenAIChatProvider,
+ policy: ReasoningPolicy,
+ expected: str,
+) -> None:
+ body = lightning_provider._chat._build_request_body(_request(), reasoning=policy)
+
+ assert body["reasoning_effort"] == expected
+
+
+async def test_extracts_catalog_metadata(
+ lightning_provider: OpenAIChatProvider,
+) -> None:
+ lightning_provider._client.models.list = AsyncMock(
+ return_value={
+ "data": [
+ _catalog_model(
+ "openai/gpt-5-mini",
+ context_length=400_000,
+ max_tokens=128_000,
+ input_modalities=("text", "image"),
+ ),
+ _catalog_model("bare-model", input_modalities=None),
+ ]
+ }
+ )
+
+ model_infos = await lightning_provider.list_model_infos()
+
+ assert model_infos == frozenset(
+ {
+ ProviderModelInfo(
+ "openai/gpt-5-mini",
+ input_modalities=frozenset(
+ {ModelInputModality.TEXT, ModelInputModality.IMAGE}
+ ),
+ context_window_tokens=400_000,
+ max_output_tokens=128_000,
+ ),
+ ProviderModelInfo("bare-model"),
+ }
+ )
+
+
+async def test_rejects_catalog_item_without_model_id(
+ lightning_provider: OpenAIChatProvider,
+) -> None:
+ item = _catalog_model()
+ item.pop("id")
+ lightning_provider._client.models.list = AsyncMock(return_value={"data": [item]})
+
+ with pytest.raises(ModelListResponseError, match="include id"):
+ await lightning_provider.list_model_infos()
+
+
+async def test_empty_catalog_rejected(lightning_provider: OpenAIChatProvider) -> None:
+ lightning_provider._client.models.list = AsyncMock(return_value={"data": []})
+
+ with pytest.raises(ModelListResponseError, match="did not include any model ids"):
+ await lightning_provider.list_model_infos()
+
+
+ "content_type", ["application/json", "text/plain; charset=utf-8"]
+)
+async def test_model_catalog_uses_documented_url_and_bearer_auth(
+ content_type: str,
+) -> None:
+ requests: list[httpx2.Request] = []
+
+ def handler(request: httpx2.Request) -> httpx2.Response:
+ requests.append(request)
+ return httpx2.Response(
+ 200,
+ headers={"content-type": content_type},
+ json={"data": [_catalog_model()]},
+ )
+
+ async with AsyncOpenAI(
+ api_key="wire-lightning-key",
+ base_url=LIGHTNING_DEFAULT_BASE,
+ max_retries=0,
+ http_client=httpx2.AsyncClient(transport=httpx2.MockTransport(handler)),
+ ) as client:
+ lightning_provider = OpenAIChatProvider(
+ make_provider_config(
+ api_key="wire-lightning-key",
+ base_url=LIGHTNING_DEFAULT_BASE,
+ ),
+ profile=OPENAI_CHAT_PROFILES["lightning"],
+ admission=immediate_admission(provider_name="lightning"),
+ client=client,
+ )
+ model_infos = await lightning_provider.list_model_infos()
+
+ assert model_infos == frozenset(
+ {
+ ProviderModelInfo(
+ _MODEL,
+ input_modalities=frozenset({ModelInputModality.TEXT}),
+ )
+ }
+ )
+ assert len(requests) == 1
+ assert str(requests[0].url) == "https://lightning.ai/api/v1/models"
+ assert requests[0].headers["authorization"] == "Bearer wire-lightning-key"
diff --git a/tests/providers/test_provider_runtime.py b/tests/providers/test_provider_runtime.py
index 7964d7e0..a9597861 100644
--- a/tests/providers/test_provider_runtime.py
+++ b/tests/providers/test_provider_runtime.py
@@ -21,6 +21,7 @@ from free_claude_code.config.provider_catalog import (
FEATHERLESS_DEFAULT_BASE,
HUGGINGFACE_DEFAULT_BASE,
KIMI_CODE_DEFAULT_BASE,
+ LIGHTNING_DEFAULT_BASE,
LLM7_DEFAULT_BASE,
MINIMAX_DEFAULT_BASE,
NARAROUTE_DEFAULT_BASE,
@@ -113,6 +114,8 @@ def _make_settings(**overrides):
mock.ollama_api_key = "test_ollama_cloud_key"
mock.poolside_api_key = "test_poolside_key"
mock.llm7_api_key = "test_llm7_key"
+ mock.lightning_api_key = "test_lightning_key"
+ mock.lightning_base_url = LIGHTNING_DEFAULT_BASE
mock.nvidia_nim_proxy = None
mock.open_router_proxy = None
mock.lmstudio_proxy = None
@@ -159,6 +162,7 @@ def _make_settings(**overrides):
mock.ollama_cloud_proxy = None
mock.poolside_proxy = None
mock.llm7_proxy = None
+ mock.lightning_proxy = None
mock.kilo_api_key = "test_kilo_key"
mock.kilo_proxy = None
mock.openai_proxy = None
@@ -927,6 +931,7 @@ async def test_create_provider_instantiates_each_builtin():
"opencode_zen": OpenCodeProvider,
"poolside": OpenAIChatProvider,
"llm7": OpenAIChatProvider,
+ "lightning": OpenAIChatProvider,
"opencode_go": OpenCodeProvider,
"vercel": OpenAIChatProvider,
"bedrock": OpenAIChatProvider,
diff --git a/uv.lock b/uv.lock
index 42764b6f..c2669028 100644
--- a/uv.lock
+++ b/uv.lock
@@ -599,7 +599,7 @@ wheels = [
[[package]]
name = "free-claude-code"
-version = "6.2.37"
+version = "6.2.38"
source = { editable = "." }
dependencies = [
{ name = "aiohttp" },