Choose a model tier or specific model for a task and understand the displayed Credit rate.
The task composer in Qoder includes a model selector. Choose the model after you understand the task scope, then check the displayed Credit rate before sending. You can use automatic routing or a performance tier, or select an available model directly.
Start with Auto for general work. Choose Ultimate or Performance for more complex work where result quality matters most, and Efficient or Lite for smaller, well-defined tasks.
The selector separates models into Default and Enterprise. Default contains the tiers and models available to the signed-in account. Enterprise contains models provided by the current organization. If the Enterprise tab is empty or unavailable, the account does not currently have an enterprise model to select.
In addition to tier selection, you can directly choose a specific model from a particular provider. This is ideal when you have a clear model preference or a specific requirement.
Some tiers and models let you tune parameters. In the model menu, point to a supported model to open its parameter panel, then select Edit.
The context window controls how much conversation and project material the model can consider in one request.
Thinking effort controls how deeply the model reasons before generating a response. Available options vary by model.
Ultimate, Performance, DeepSeek-V4-Pro, DeepSeek-V4-Flash, Qwen3.8-Max, Qwen3.8-Flash, Qwen3.7-Max, Qwen3.7-Plus, GLM-5.3, MiniMax-M3, and Kimi-K3 support parameter configuration. Qwen3.7-Max, Qwen3.7-Plus, and MiniMax-M3 support Context only; the others support both Context and Thinking Effort.
Select Model settings at the bottom of the selector when you need to manage the models available to the app. Settings and parameters vary by model; only use controls that appear for the selected model.
Yes. The new selection applies to later messages and keeps the current task context; earlier responses are not regenerated.
Models vary by client version, account availability, and service updates. The names, labels, and rates shown in the composer are authoritative for the current task.
A tier lets Qoder route by policy. Select a specific model when you explicitly need that model or capability combination.
Credit usage depends on the tokens used, the selected tier or model, configured parameters, and tool activity. Check the rate shown in the selector before sending.
Only Lite is available until the account is upgraded or receives more Credits.
To use your own API key, add a personal model under Settings → Models. After it is added, the model becomes available to task, Agent, and automation model selectors. See Custom Models.
Choose a tier
| Tier | Description | Use cases | Credit usage |
|---|---|---|---|
| Auto (Smart Routing) | Intelligently selects the most suitable model, balancing performance and cost | Most daily work, recommended as default | ~1.0× |
| Ultimate | Expert-level deep reasoning and thinking capabilities | Complex system design, high-difficulty problem analysis | ~1.6× |
| Performance | Advanced reasoning capabilities, high-quality output | Core feature implementation, architecture design, code refactoring | ~1.1× |
| Efficient | Standard reasoning capabilities, high cost-effectiveness | Basic code generation, unit tests, daily Q&A | ~0.3× |
| Lite | Basic reasoning capabilities, free to use | Quick validation, basic logic implementation, quick Q&A | Free |
Credit consumption comparison
| Model tier | Credit consumption rate | Example task consumption |
|---|---|---|
| Auto | ~1.0× | 10 Credits |
| Ultimate | ~1.6× | 20 Credits |
| Performance | ~1.1× | 11 Credits |
| Efficient | ~0.3× | 3 Credits |
| Lite | Free | 0 Credits |
Due to variations in tasks and codebases, actual consumption rates may differ.Actual Credit use varies with the task, repository, selected parameters, and tool calls. The rate shown in the selector is the authoritative estimate for the next request.
Default and Enterprise models
The selector separates models into Default and Enterprise. Default contains the tiers and models available to the signed-in account. Enterprise contains models provided by the current organization. If the Enterprise tab is empty or unavailable, the account does not currently have an enterprise model to select.
Select a specific model
In addition to tier selection, you can directly choose a specific model from a particular provider. This is ideal when you have a clear model preference or a specific requirement.
| Model name | Description | Credit consumption rate |
|---|---|---|
| Qwen3.8-Max | Qwen's latest foundation model, with 2.4T parameters, leading in software engineering, office work, and deep reasoning | 0.5× |
| Qwen3.8-Flash | Qwen's open-weight multimodal MoE model, delivering an excellent balance of capability, latency, and cost | 0.1× |
| Qwen3.7-Max | Qwen's latest model, with top-tier agentic capabilities, autonomously handles complex tasks up to 35 hours long | 0.5× |
| Qwen3.7-Plus | Comprehensive leap in reasoning capability, efficiency, and multimodal experience | 0.1× |
| DeepSeek-V4-Pro | Excels at complex reasoning, code generation, and engineering tasks | 0.8× |
| DeepSeek-V4-Flash | Fast reasoning and low cost with balanced capabilities | 0.3× |
| GLM-5.3 | Zhipu's open-source flagship model, with coding capabilities competitive with leading international models, plus security review and vulnerability reasoning | 0.6× |
| Kimi-K3 | Kimi's most powerful model, with 2.8T parameters, built for software engineering, knowledge work, and deep reasoning | 0.8× |
| Kimi-K2.7-Code | Kimi's latest model, purpose-built for long-context coding: precise instruction following, reliable long-horizon execution | 0.3× |
| MiniMax-M3 | Native multimodal perception, frontier coding, and 1M context depth for highly demanding workflows | 0.2× |
Models, capability labels, and rates can change by client version and account availability. The selector in the task composer is authoritative.
Configure model parameters
Some tiers and models let you tune parameters. In the model menu, point to a supported model to open its parameter panel, then select Edit.
Context window
The context window controls how much conversation and project material the model can consider in one request.
| Option | Description |
|---|---|
| 200K | Standard context window, sufficient for most tasks |
| 400K | Extended context for larger codebases or lengthy conversations |
| 1M | Maximum context for extremely large-scale projects |
Thinking effort
Thinking effort controls how deeply the model reasons before generating a response. Available options vary by model.
| Option | Description |
|---|---|
| low | Minimal reasoning, fastest responses |
| medium | Balanced reasoning depth |
| high | Thorough reasoning for complex tasks |
| xhigh | Deep analysis for high-difficulty problems |
| max | Maximum reasoning depth, best for the most complex challenges |
Open model settings
Select Model settings at the bottom of the selector when you need to manage the models available to the app. Settings and parameters vary by model; only use controls that appear for the selected model.
Switch models
1
Open the selector
Select the current model at the bottom of a new or existing task composer.
2
Choose a tier or model
Select Auto, another performance tier, or a specific model from the list.
3
Continue the task
The selection applies to subsequent messages. Before sending, check the model and rate shown in the composer.