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

Feature/poolside support (#1483)

Parsa.py20b7e45f88c4
diff --git a/.env.example b/.env.example
index afdaf61f..d908ca4f 100644
--- a/.env.example
+++ b/.env.example
@@ -78,6 +78,10 @@ TOKENROUTER_BASE_URL="https://api.tokenrouter.com/v1"
# NARAROUTE_API_KEY=<your-token>
NARAROUTE_BASE_URL="https://router.bynara.id/v1"
+
+# Poolside AI (OpenAI-compatible Chat Completions at inference.poolside.ai/v1)
+# POOLSIDE_API_KEY=<your-token>
+
# Fireworks AI Config (OpenAI-compatible Chat Completions at api.fireworks.ai/inference/v1)
# FIREWORKS_API_KEY=<your-token>
@@ -187,7 +191,7 @@ 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" | "github_models" | "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"
+# Valid providers: "nvidia_nim" | "openai" | "azure_openai" | "open_router" | "gemini" | "vertex" | "deepseek" | "mistral" | "mistral_codestral" | "opencode_zen" | "opencode_go" | "vercel" | "bedrock" | "huggingface" | "cohere" | "github_models" | "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"
# MODEL_FABLE=<provider/model-id>
# MODEL_OPUS=<provider/model-id>
# MODEL_SONNET=<provider/model-id>
@@ -227,6 +231,7 @@ MODEL="nvidia_nim/nvidia/nemotron-3-super-120b-a12b"
# FCC_SMOKE_MODEL_ZAI_API=<provider/model-id>
# FCC_SMOKE_MODEL_TOKENROUTER=<your-token>
# FCC_SMOKE_MODEL_NARAROUTE=<provider/model-id>
+# FCC_SMOKE_MODEL_POOLSIDE=<provider/model-id>
# FCC_SMOKE_MODEL_FIREWORKS=<provider/model-id>
# FCC_SMOKE_MODEL_NOVITA=<provider/model-id>
# FCC_SMOKE_MODEL_CLOUDFLARE=<provider/model-id>
@@ -302,6 +307,7 @@ REASONING_HAIKU=inherit
# ZAI_API_PROXY=<http-or-socks-url>
# TOKENROUTER_PROXY=<http-or-socks-url>
# NARAROUTE_PROXY=<http-or-socks-url>
+# POOLSIDE_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 279b2e1e..e5614151 100644
--- a/README.md
+++ b/README.md
@@ -25,7 +25,7 @@
## What You Get
-- **48 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.
+- **49 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.
- **9 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), or [Muse Code](https://research.meta.ai/blog/introducing-muse-code-and-muse-spark-1-2/) 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.
@@ -249,6 +249,7 @@ from more than one provider before succeeding.
| [Z.ai API (pay as you go)](https://z.ai/manage-apikey/apikey-list) | `ZAI_API_KEY` | `zai_api/glm-4.7-flash` |
| [TokenRouter](https://www.tokenrouter.com/) | `TOKENROUTER_API_KEY` | `tokenrouter/moonshotai/kimi-k3-free` |
| [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` |
| [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 f850d0a5..7fb27a74 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
[project]
name = "free-claude-code"
-version = "5.12.2"
+version = "5.13.0"
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 ea4eeaa5..87675620 100644
--- a/smoke/lib/config.py
+++ b/smoke/lib/config.py
@@ -89,6 +89,7 @@ PROVIDER_SMOKE_DEFAULT_MODELS: dict[str, str] = {
"cloudflare": "cloudflare/@cf/moonshotai/kimi-k2.6",
"tokenrouter": "tokenrouter/moonshotai/kimi-k3-free",
"nararoute": "nararoute/kimi-k3-free",
+ "poolside": "poolside/poolside/laguna-s-2.1",
"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/config/provider_catalog.py b/src/free_claude_code/config/provider_catalog.py
index 4d27354b..bbf3c2a3 100644
--- a/src/free_claude_code/config/provider_catalog.py
+++ b/src/free_claude_code/config/provider_catalog.py
@@ -77,6 +77,8 @@ FEATHERLESS_DEFAULT_BASE = "https://api.featherless.ai/v1"
TOKENROUTER_DEFAULT_BASE = "https://api.tokenrouter.com/v1"
# NaraRoute OpenAI-compatible Chat Completions gateway.
NARAROUTE_DEFAULT_BASE = "https://router.bynara.id/v1"
+# Poolside AI OpenAI-compatible Chat Completions API.
+POOLSIDE_DEFAULT_BASE = "https://inference.poolside.ai/v1"
# Agnes AI OpenAI-compatible Chat Completions API.
AGNES_DEFAULT_BASE = "https://apihub.agnes-ai.com/v1"
# ZenMux OpenAI-compatible Chat Completions gateway.
@@ -527,6 +529,15 @@ PROVIDER_CATALOG: dict[str, ProviderDescriptor] = {
base_url_attr="nararoute_base_url",
proxy_attr="nararoute_proxy",
),
+ "poolside": ProviderDescriptor(
+ provider_id="poolside",
+ display_name="Poolside AI",
+ credential_env="POOLSIDE_API_KEY",
+ credential_url="https://platform.poolside.ai/",
+ credential_attr="poolside_api_key",
+ default_base_url=POOLSIDE_DEFAULT_BASE,
+ proxy_attr="poolside_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 7f2f814c..d0b605fb 100644
--- a/src/free_claude_code/config/settings.py
+++ b/src/free_claude_code/config/settings.py
@@ -196,6 +196,11 @@ class Settings(BaseModel):
validation_alias="NARAROUTE_BASE_URL",
)
+ # ==================== Poolside AI (OpenAI-compatible) ====================
+ poolside_api_key: OptionalNonEmptyString = Field(
+ default=None, validation_alias="POOLSIDE_API_KEY"
+ )
+
# ==================== Fireworks AI Config ====================
fireworks_api_key: OptionalNonEmptyString = Field(
default=None, validation_alias="FIREWORKS_API_KEY"
@@ -482,6 +487,9 @@ class Settings(BaseModel):
nararoute_proxy: OptionalNonEmptyString = Field(
default=None, validation_alias="NARAROUTE_PROXY"
)
+ poolside_proxy: OptionalNonEmptyString = Field(
+ default=None, validation_alias="POOLSIDE_PROXY"
+ )
fireworks_proxy: OptionalNonEmptyString = Field(
default=None, validation_alias="FIREWORKS_PROXY"
)
@@ -509,7 +517,6 @@ class Settings(BaseModel):
ollama_cloud_proxy: OptionalNonEmptyString = Field(
default=None, validation_alias="OLLAMA_CLOUD_PROXY"
)
-
# ==================== Provider Rate Limiting ====================
provider_rate_limit: int = Field(default=1, validation_alias="PROVIDER_RATE_LIMIT")
provider_rate_window: int = Field(
diff --git a/src/free_claude_code/providers/openai_chat/profiles.py b/src/free_claude_code/providers/openai_chat/profiles.py
index 6dac8048..eaa8807e 100644
--- a/src/free_claude_code/providers/openai_chat/profiles.py
+++ b/src/free_claude_code/providers/openai_chat/profiles.py
@@ -613,6 +613,16 @@ OPENAI_CHAT_PROFILES: dict[str, OpenAIChatProfile] = {
enabled_value="medium",
),
),
+ "poolside": OpenAIChatProfile(
+ _policy(
+ "POOLSIDE",
+ ReasoningReplayMode.REASONING_CONTENT,
+ include_extra_body=True,
+ extra_body_validator=validate_extra_body_does_not_override_reasoning_fields,
+ default_max_tokens=ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS,
+ ),
+ ChatTemplateReasoning(field="enable_thinking"),
+ ),
"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 50921604..3f214576 100644
--- a/tests/contracts/test_provider_catalog_order.py
+++ b/tests/contracts/test_provider_catalog_order.py
@@ -50,6 +50,7 @@ _EXPECTED_PROVIDER_ORDER: tuple[str, ...] = (
"zai_api",
"tokenrouter",
"nararoute",
+ "poolside",
"ollama_cloud",
"lmstudio",
"llamacpp",
diff --git a/tests/contracts/test_smoke_config.py b/tests/contracts/test_smoke_config.py
index 7a455cda..d18003d1 100644
--- a/tests/contracts/test_smoke_config.py
+++ b/tests/contracts/test_smoke_config.py
@@ -67,6 +67,7 @@ def _settings(**overrides):
"sambanova_api_key": "",
"cerebras_api_key": "",
"ollama_api_key": "",
+ "poolside_api_key": "",
"fireworks_api_key": "",
"novita_api_key": "",
"cloudflare_api_token": "",
@@ -162,6 +163,25 @@ def test_provider_smoke_models_cover_configured_providers_independent_of_model_m
assert models[0].source == "provider_default"
+def test_poolside_provider_configuration_uses_documented_default_model(
+ monkeypatch,
+) -> None:
+ monkeypatch.delenv("FCC_SMOKE_MODEL_POOLSIDE", raising=False)
+ config = _smoke_config(
+ settings=_settings(
+ model="ollama/llama3.1",
+ ollama_base_url="",
+ poolside_api_key="poolside-key",
+ )
+ )
+
+ assert config.has_provider_configuration("poolside")
+ models = config.provider_smoke_models()
+ assert [model.provider for model in models] == ["poolside"]
+ assert models[0].full_model == "poolside/poolside/laguna-s-2.1"
+ assert models[0].source == "provider_default"
+
+
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_poolside.py b/tests/providers/test_poolside.py
new file mode 100644
index 00000000..4c81fa4c
--- /dev/null
+++ b/tests/providers/test_poolside.py
@@ -0,0 +1,217 @@
+"""Tests for the Poolside AI OpenAI-chat provider profile."""
+
+from types import SimpleNamespace
+
+import httpx2
+import pytest
+from openai import AsyncOpenAI
+
+from free_claude_code.application.errors import InvalidRequestError
+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 POOLSIDE_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.reasoning import ReasoningPolicy
+from free_claude_code.providers.openai_chat import OpenAIChatProvider
+from tests.providers.support import (
+ REASONING_OFF,
+ REASONING_ON,
+ capture_openai_chat_wire_body,
+ immediate_admission,
+ make_provider_config,
+ profiled_provider,
+ reasoning_for,
+)
+
+_MODEL = "poolside/laguna-s-2.1"
+
+
+def poolside_provider() -> OpenAIChatProvider:
+ return profiled_provider(
+ "poolside",
+ make_provider_config(
+ api_key="test-poolside-key",
+ base_url=POOLSIDE_DEFAULT_BASE,
+ rate_limit=10,
+ rate_window=60,
+ ),
+ admission=immediate_admission(provider_name="poolside", max_attempts=1),
+ )
+
+
+def _request(**overrides: JsonValue) -> MessagesRequest:
+ payload: JsonObject = {
+ "model": _MODEL,
+ "messages": [{"role": "user", "content": "Hello"}],
+ }
+ payload.update(overrides)
+ return MessagesRequest.model_validate(payload)
+
+
+ ("reasoning", "enabled"),
+ [(REASONING_OFF, False), (REASONING_ON, True)],
+)
+def test_encodes_documented_chat_template_thinking_control(
+ poolside_provider: OpenAIChatProvider,
+ reasoning: ReasoningPolicy,
+ enabled: bool,
+) -> None:
+ body = poolside_provider._build_request_body(
+ _request(extra_body={"top_k": 20}),
+ reasoning=reasoning,
+ )
+
+ assert body["max_tokens"] == ANTHROPIC_DEFAULT_MAX_OUTPUT_TOKENS
+ assert body["extra_body"] == {
+ "top_k": 20,
+ "chat_template_kwargs": {"enable_thinking": enabled},
+ }
+
+
+def test_omits_thinking_control_for_poolside_default(
+ poolside_provider: OpenAIChatProvider,
+) -> None:
+ body = poolside_provider._build_request_body(
+ _request(),
+ reasoning=ReasoningPolicy.provider_default(),
+ )
+
+ assert "extra_body" not in body
+
+
+ "field",
+ ["reasoning", "reasoning_effort", "thinking", "enable_thinking"],
+)
+def test_rejects_caller_reasoning_override(
+ poolside_provider: OpenAIChatProvider,
+ field: str,
+) -> None:
+ request = _request(extra_body={field: "caller-owned"})
+
+ with pytest.raises(InvalidRequestError, match="must not override reasoning"):
+ poolside_provider._build_request_body(request, reasoning=REASONING_ON)
+
+
+def test_replays_reasoning_content_with_tool_history(
+ poolside_provider: OpenAIChatProvider,
+) -> None:
+ request = _request(
+ messages=[
+ {"role": "user", "content": "Inspect the file."},
+ {
+ "role": "assistant",
+ "content": [
+ {"type": "thinking", "thinking": "Read it first."},
+ {"type": "text", "text": "I will inspect it."},
+ {
+ "type": "tool_use",
+ "id": "toolu_1",
+ "name": "read_file",
+ "input": {"path": "example.py"},
+ },
+ ],
+ },
+ {
+ "role": "user",
+ "content": [
+ {
+ "type": "tool_result",
+ "tool_use_id": "toolu_1",
+ "content": "print('hello')",
+ }
+ ],
+ },
+ ]
+ )
+
+ body = poolside_provider._build_request_body(
+ request,
+ reasoning=reasoning_for(request),
+ )
+
+ assert body["messages"][1] == {
+ "role": "assistant",
+ "content": "I will inspect it.",
+ "reasoning_content": "Read it first.",
+ "tool_calls": [
+ {
+ "id": "toolu_1",
+ "type": "function",
+ "function": {
+ "name": "read_file",
+ "arguments": '{"path": "example.py"}',
+ },
+ }
+ ],
+ }
+ assert body["messages"][2] == {
+ "role": "tool",
+ "tool_call_id": "toolu_1",
+ "content": "print('hello')",
+ }
+ assert (
+ poolside_provider._profile.reasoning_delta(
+ SimpleNamespace(reasoning_content="next thought")
+ )
+ == "next thought"
+ )
+
+
+async def test_thinking_control_reaches_wire_as_top_level_extension(
+ poolside_provider: OpenAIChatProvider,
+) -> None:
+ body = poolside_provider._build_request_body(
+ _request(),
+ reasoning=REASONING_OFF,
+ )
+
+ wire = await capture_openai_chat_wire_body(body)
+
+ assert wire["chat_template_kwargs"] == {"enable_thinking": False}
+ assert "extra_body" not in wire
+
+
+async def test_model_catalog_uses_documented_url_and_bearer_auth(
+ poolside_provider: OpenAIChatProvider,
+) -> None:
+ requests: list[httpx2.Request] = []
+
+ def handler(request: httpx2.Request) -> httpx2.Response:
+ requests.append(request)
+ return httpx2.Response(
+ 200,
+ json={
+ "object": "list",
+ "data": [
+ {
+ "id": _MODEL,
+ "object": "model",
+ "created": 0,
+ "owned_by": "poolside",
+ }
+ ],
+ },
+ )
+
+ await poolside_provider._client.close()
+ poolside_provider._client = AsyncOpenAI(
+ api_key="wire-poolside-key",
+ base_url=POOLSIDE_DEFAULT_BASE,
+ max_retries=0,
+ http_client=httpx2.AsyncClient(transport=httpx2.MockTransport(handler)),
+ )
+ try:
+ model_infos = await poolside_provider.list_model_infos()
+ finally:
+ await poolside_provider.cleanup()
+
+ assert model_infos == frozenset({ProviderModelInfo(_MODEL)})
+ assert len(requests) == 1
+ assert str(requests[0].url) == "https://inference.poolside.ai/v1/models"
+ assert requests[0].headers["authorization"] == "Bearer wire-poolside-key"
diff --git a/tests/providers/test_provider_runtime.py b/tests/providers/test_provider_runtime.py
index 63e278f2..2062dd0e 100644
--- a/tests/providers/test_provider_runtime.py
+++ b/tests/providers/test_provider_runtime.py
@@ -25,6 +25,7 @@ from free_claude_code.config.provider_catalog import (
NARAROUTE_DEFAULT_BASE,
NEBIUS_DEFAULT_BASE,
OLLAMA_CLOUD_DEFAULT_BASE,
+ POOLSIDE_DEFAULT_BASE,
PROVIDER_CATALOG,
QWENCLOUD_CODING_DEFAULT_BASE,
QWENCLOUD_DEFAULT_BASE,
@@ -107,6 +108,7 @@ def _make_settings(**overrides):
mock.llamacpp_base_url = "http://localhost:8080/v1"
mock.ollama_base_url = "http://localhost:11434"
mock.ollama_api_key = "test_ollama_cloud_key"
+ mock.poolside_api_key = "test_poolside_key"
mock.nvidia_nim_proxy = None
mock.open_router_proxy = None
mock.lmstudio_proxy = None
@@ -152,6 +154,7 @@ def _make_settings(**overrides):
mock.cerebras_api_key = ""
mock.cerebras_proxy = None
mock.ollama_cloud_proxy = None
+ mock.poolside_proxy = None
mock.kilo_api_key = "test_kilo_key"
mock.kilo_proxy = None
mock.openai_proxy = None
@@ -233,6 +236,30 @@ def test_ollama_cloud_provider_config_uses_key_and_proxy():
assert config.proxy == "http://proxy.test:8080"
+def test_poolside_provider_config_uses_key_base_and_proxy() -> None:
+ descriptor = PROVIDER_CATALOG["poolside"]
+ settings = _make_settings(
+ poolside_api_key="poolside-token",
+ poolside_proxy="http://proxy.test:8080",
+ )
+
+ config = build_provider_config(descriptor, settings)
+ with patch("free_claude_code.providers.openai_chat.provider.AsyncOpenAI"):
+ provider = create_provider("poolside", settings)
+
+ assert descriptor.display_name == "Poolside AI"
+ assert descriptor.credential_env == "POOLSIDE_API_KEY"
+ assert descriptor.credential_attr == "poolside_api_key"
+ assert descriptor.credential_url == "https://platform.poolside.ai/"
+ assert descriptor.default_base_url == POOLSIDE_DEFAULT_BASE
+ assert descriptor.base_url_attr is None
+ assert descriptor.proxy_attr == "poolside_proxy"
+ assert config.api_key == "poolside-token"
+ assert config.base_url == POOLSIDE_DEFAULT_BASE
+ assert config.proxy == "http://proxy.test:8080"
+ assert isinstance(provider, OpenAIChatProvider)
+
+
def test_xai_provider_config_uses_key_base_and_proxy() -> None:
descriptor = PROVIDER_CATALOG["xai"]
settings = _make_settings(
@@ -843,6 +870,7 @@ def test_create_provider_instantiates_each_builtin():
"ollama_cloud": OpenAIChatProvider,
"wafer": OpenAIChatProvider,
"opencode_zen": OpenAIChatProvider,
+ "poolside": OpenAIChatProvider,
"opencode_go": OpenAIChatProvider,
"vercel": OpenAIChatProvider,
"bedrock": OpenAIChatProvider,
diff --git a/uv.lock b/uv.lock
index 72df5aa8..eefb9276 100644
--- a/uv.lock
+++ b/uv.lock
@@ -607,7 +607,7 @@ wheels = [
[[package]]
name = "free-claude-code"
-version = "5.12.2"
+version = "5.13.0"
source = { editable = "." }
dependencies = [
{ name = "aiohttp" },