Add zai-org/GLM-4.5-FP8 model to Chutes AI provider

[leakless21]

What specific problem does this solve?

The Roo Code extension currently lacks support for the zai-org/GLM-4.5-FP8 model on the Chutes AI provider, which offers 128k token context window. This model performs on par with SOTA models while being much faster and cheaper.

Impact: All Roo Code users who needs efficient model.

Additional context (optional)

GLM-4.5 is Zai's latest flagship foundation model, purpose-built for agent-based applications. It leverages a Mixture-of-Experts (MoE) architecture and supports a context length of up to 128k tokens. GLM-4.5 delivers significantly enhanced capabilities in reasoning, code generation, and agent alignment. It supports a hybrid inference mode with two options, a "thinking mode" designed for complex reasoning and tool use, and a "non-thinking mode" optimized for instant responses.

Roo Code Task Links (Optional)

No response

Request checklist

• I've searched existing Issues and Discussions for duplicates
• This describes a specific problem with clear impact and context

Interested in implementing this?

• Yes, I'd like to help implement this feature

Implementation requirements

• I understand this needs approval before implementation begins

How should this be solved? (REQUIRED if contributing, optional otherwise)

No response

How will we know it works? (Acceptance Criteria - REQUIRED if contributing, optional otherwise)

No response

Technical considerations (REQUIRED if contributing, optional otherwise)

No response

Trade-offs and risks (REQUIRED if contributing, optional otherwise)

No response

[roomote]

I'm here to help out the maintainers and am going to see if I can fix this issue or at least get close enough to be helpful! I'll work on adding the zai-org/GLM-4.5-FP8 model to the Chutes AI provider. Let me explore the codebase to understand how models are configured and implement this enhancement.

[mrubens]

@leakless21 is that significantly different than the GLM-4.5-Air that we recently added?

[roomote]

✅ Implementation Complete!

I've successfully implemented the requested zai-org/GLM-4.5-FP8 model support for the Chutes AI provider.

🔗 Pull Request

PR #6441: #6441

📋 What was implemented:

✅ Model Type Definition: Added zai-org/GLM-4.5-FP8 to the ChutesModelId type
✅ Model Configuration: Configured with 128k context window (131072 tokens) as requested
✅ Test Coverage: Added comprehensive test case to ensure proper functionality
✅ Quality Assurance: All tests pass, linting passes, type checking passes

🚀 Model Features:

• 128k token context window - Perfect for complex agent-based applications
• MoE architecture - Optimized for performance and efficiency
• Agent-optimized - Built specifically for reasoning, code generation, and agent alignment
• Hybrid inference modes - Supports both thinking and non-thinking modes

The PR is currently going through CI checks, and several have already passed. Once all checks complete and the PR is reviewed/merged, the GLM-4.5-FP8 model will be available for all Roo Code users through the Chutes AI provider! 🎉

[leakless21]

Yes, it has 355b parameters compares to 106b of Air variant. Also the benchmark results are better, especially at agentic and coding uses.

…

On Wed, Jul 30, 2025, 22:29 Matt Rubens ***@***.***> wrote: *mrubens* left a comment (RooCodeInc/Roo-Code#6440) <#6440 (comment)> @leakless21 <https://github.com/leakless21> is that significantly different than the GLM-4.5-Air that we recently added? — Reply to this email directly, view it on GitHub <#6440 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BANWVTUVEZZPZQJTLE5P2DT3LDQEHAVCNFSM6AAAAACCXJDJASVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCMZWHA2DAMRTG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>