feat: add MiniMax as alternative LLM provider for Bot#1560
Open
octo-patch wants to merge 1 commit into
Open
Conversation
Add MiniMax (MiniMax-M2.7, MiniMax-M2.5, MiniMax-M2.5-highspeed) as an alternative LLM provider alongside OpenAI for the Bot abstraction. MiniMax uses an OpenAI-compatible API, so this is achieved by adding a configurable base_url to BotConfig and auto-detecting the provider from the model name. Changes: - Go backend: Add BaseUrl field to BotConfig, use openai.DefaultConfig() + NewClientWithConfig() when base URL is provided - Python SDK: Add MiniMax models to VALID_MODELS, add base_url parameter with auto-detection for MiniMax models - Tests: 8 Go unit tests + 21 Python unit/integration tests - README: Add multi-provider LLM support to features list
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
3 issues found across 7 files
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="sdk/tests/test_bot_standalone.py">
<violation number="1" location="sdk/tests/test_bot_standalone.py:197">
P2: Broad `except Exception` in integration tests converts assertion failures into skips, masking real regressions.</violation>
</file>
<file name="sdk/tests/test_bot.py">
<violation number="1" location="sdk/tests/test_bot.py:56">
P1: Blanket `except Exception: pass` in tests suppresses assertion/runtime failures, causing false-positive passing tests and masking provider-regression bugs.</violation>
</file>
<file name="sdk/src/beta9/abstractions/experimental/bot/bot.py">
<violation number="1" location="sdk/src/beta9/abstractions/experimental/bot/bot.py:231">
P2: The new “OpenAI-compatible API” support is blocked by the existing model allowlist: non-allowlisted models still raise ValueError before base_url is applied, so custom providers cannot be used despite the added documentation.</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
| except ValueError as e: | ||
| if "Invalid model name" in str(e): | ||
| pytest.fail(f"Model {model} should be accepted but was rejected") | ||
| except Exception: |
Contributor
There was a problem hiding this comment.
P1: Blanket except Exception: pass in tests suppresses assertion/runtime failures, causing false-positive passing tests and masking provider-regression bugs.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At sdk/tests/test_bot.py, line 56:
<comment>Blanket `except Exception: pass` in tests suppresses assertion/runtime failures, causing false-positive passing tests and masking provider-regression bugs.</comment>
<file context>
@@ -0,0 +1,181 @@
+ except ValueError as e:
+ if "Invalid model name" in str(e):
+ pytest.fail(f"Model {model} should be accepted but was rejected")
+ except Exception:
+ # Other errors (gRPC, network) are expected in test environment
+ pass
</file context>
| try: | ||
| resp = urllib.request.urlopen(req, timeout=10) | ||
| self.assertEqual(resp.status, 200) | ||
| except Exception: |
Contributor
There was a problem hiding this comment.
P2: Broad except Exception in integration tests converts assertion failures into skips, masking real regressions.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At sdk/tests/test_bot_standalone.py, line 197:
<comment>Broad `except Exception` in integration tests converts assertion failures into skips, masking real regressions.</comment>
<file context>
@@ -0,0 +1,267 @@
+ try:
+ resp = urllib.request.urlopen(req, timeout=10)
+ self.assertEqual(resp.status, 200)
+ except Exception:
+ self.skipTest("MiniMax API not reachable")
+
</file context>
| api_key (str): | ||
| OpenAI API key to use for the bot. In the future this will support other LLM providers. | ||
| API key for the LLM provider. Works with OpenAI, MiniMax, or any | ||
| OpenAI-compatible API. |
Contributor
There was a problem hiding this comment.
P2: The new “OpenAI-compatible API” support is blocked by the existing model allowlist: non-allowlisted models still raise ValueError before base_url is applied, so custom providers cannot be used despite the added documentation.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At sdk/src/beta9/abstractions/experimental/bot/bot.py, line 231:
<comment>The new “OpenAI-compatible API” support is blocked by the existing model allowlist: non-allowlisted models still raise ValueError before base_url is applied, so custom providers cannot be used despite the added documentation.</comment>
<file context>
@@ -224,8 +224,14 @@ class Bot(RunnerAbstraction, DeployableMixin):
api_key (str):
- OpenAI API key to use for the bot. In the future this will support other LLM providers.
+ API key for the LLM provider. Works with OpenAI, MiniMax, or any
+ OpenAI-compatible API.
+ base_url (Optional[str]):
+ Custom base URL for the LLM API. When using MiniMax models, this
</file context>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
base_urlonBotConfig-- zero new dependencies requiredChanges
Go Backend (
pkg/abstractions/experimental/bot/)BaseUrlfield toBotConfig(omitempty for backward compatibility)openai.DefaultConfig()+NewClientWithConfig()whenBaseUrlis set, falls back toopenai.NewClient()for OpenAIPython SDK (
sdk/src/beta9/abstractions/experimental/bot/)VALID_MODELS, addbase_urlparameter with auto-detection for MiniMax models, addPROVIDER_BASE_URLSdictTests
Documentation
Usage
Test plan
Summary by cubic
Add MiniMax as an alternative LLM provider for the
Bot, with automatic provider detection and OpenAI‑compatiblebase_urlsupport. Existing OpenAI behavior is unchanged; no new dependencies.BaseUrltoBotConfig; useopenai.NewClientWithConfigwhen set.base_urlparam; auto‑set MiniMax URL forMiniMax-*; extendVALID_MODELS.Written for commit 16ae47e. Summary will update on new commits.