Skip to content

Commit c6e632b

Browse files
mrubensmrubens
authored
Add docs for baseten (#428)
1 parent c9224d7 commit c6e632b

File tree

2 files changed

+123
-0
lines changed

2 files changed

+123
-0
lines changed

‎docs/providers/baseten.md‎

Lines changed: 122 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,122 @@
1+
---
2+
title: Baseten
3+
sidebar_label: Baseten
4+
description: Learn how to configure and use Baseten's Model APIs with Roo Code. Access frontier open-source models with enterprise-grade performance, reliability, and competitive pricing.
5+
keywords:
6+
- Baseten
7+
- Model APIs
8+
- open-source models
9+
- DeepSeek
10+
- Kimi K2
11+
- Qwen
12+
- Roo Code
13+
- AI integration
14+
- API key
15+
- enterprise inference
16+
image: /img/social-share.jpg
17+
---
18+
19+
# Using Baseten With Roo Code
20+
21+
Baseten provides on-demand frontier model APIs designed for production applications, not just experimentation. Built on the Baseten Inference Stack, these APIs deliver optimized inference for leading open-source models from OpenAI, DeepSeek, Moonshot AI, and Alibaba Cloud.
22+
23+
**Website:** [https://www.baseten.co/products/model-apis/](https://www.baseten.co/products/model-apis/)
24+
25+
---
26+
27+
## Getting an API Key
28+
29+
1. **Sign Up/Sign In:** Go to [Baseten](https://www.baseten.co/) and create an account or sign in.
30+
31+
2. **Navigate to API Keys:** Access your dashboard and go to the API Keys section at [https://app.baseten.co/settings/api_keys](https://app.baseten.co/settings/api_keys).
32+
33+
3. **Create a Key:** Generate a new API key. Give it a descriptive name (e.g., "Roo Code").
34+
35+
4. **Copy the Key:** Copy the API key immediately and store it securely.
36+
37+
---
38+
39+
## Supported Models
40+
41+
Roo Code supports all current models under Baseten Model APIs, including:
42+
43+
| Model | Provider | Context | Input/Output per 1M tokens |
44+
|-------|----------|---------|---------------------------|
45+
| `moonshotai/Kimi-K2-Thinking` | Moonshot AI | 262K | $0.60 / $2.50 |
46+
| `zai-org/GLM-4.6` | Z AI | 200K | $0.60 / $2.20 |
47+
| `moonshotai/Kimi-K2-Instruct-0905` | Moonshot AI | 262K | $0.60 / $2.50 |
48+
| `openai/gpt-oss-120b` | OpenAI | 128K | $0.10 / $0.50 |
49+
| `Qwen/Qwen3-Coder-480B-A35B-Instruct` | Alibaba Cloud | 262K | $0.38 / $1.53 |
50+
| `Qwen/Qwen3-235B-A22B-Instruct-2507` | Alibaba Cloud | 262K | $0.22 / $0.80 |
51+
| `deepseek-ai/DeepSeek-R1` | DeepSeek | 163K | $2.55 / $5.95 |
52+
| `deepseek-ai/DeepSeek-R1-0528` | DeepSeek | 163K | $2.55 / $5.95 |
53+
| `deepseek-ai/DeepSeek-V3.1` | DeepSeek | 163K | $0.50 / $1.50 |
54+
| `deepseek-ai/DeepSeek-V3-0324` | DeepSeek | 163K | $0.77 / $0.77 |
55+
56+
For the most up-to-date pricing, visit the [Baseten Model APIs page](https://www.baseten.co/products/model-apis/).
57+
58+
---
59+
60+
## Configuration in Roo Code
61+
62+
1. **Open Roo Code Settings:** Click the gear icon (<Codicon name="gear" />) in the Roo Code panel.
63+
64+
2. **Select Provider:** Choose "Baseten" from the "API Provider" dropdown.
65+
66+
3. **Enter API Key:** Paste your Baseten API key into the "Baseten API Key" field.
67+
68+
4. **Select Model:** Choose your desired model from the "Model" dropdown.
69+
70+
:::warning Kimi K2 Thinking Model
71+
To use the `moonshotai/Kimi-K2-Thinking` model, you must enable native tool calling in the Roo Code settings. This setting allows Roo Code to call the model's tools through their native tool processor and is required for this reasoning model to function properly.
72+
:::
73+
74+
---
75+
76+
## Production-First Architecture
77+
78+
Baseten's Model APIs are built for production environments with several key advantages:
79+
80+
### Enterprise-Grade Reliability
81+
82+
- **Four nines of uptime** (99.99%) through active-active redundancy
83+
- **Cloud-agnostic, multi-cluster autoscaling** for consistent availability
84+
- **SOC 2 Type II certified** and **HIPAA compliant** for security requirements
85+
86+
### Optimized Performance
87+
88+
- **Pre-optimized models** shipped with the Baseten Inference Stack
89+
- **Latest-generation GPUs** with multi-cloud infrastructure
90+
- **Ultra-fast inference** optimized from the bottom up for production workloads
91+
92+
### Cost Efficiency
93+
94+
- **5-10x less expensive** than closed alternatives
95+
- **Optimized multi-cloud infrastructure** for efficient resource utilization
96+
- **Transparent pricing** with no hidden costs or rate limit surprises
97+
98+
### Developer Experience
99+
100+
- **OpenAI compatible API** - migrate by swapping a single URL
101+
- **Drop-in replacement** for closed models with comprehensive observability and analytics
102+
- **Seamless scaling** from Model APIs to dedicated deployments
103+
104+
---
105+
106+
## Special Features
107+
108+
### Function Calling & Tool Use
109+
110+
All Baseten models support structured outputs, function calling, and tool use as part of the Baseten Inference Stack, making them ideal for agentic applications and coding workflows.
111+
112+
---
113+
114+
## Tips and Notes
115+
116+
- **Static Model List:** Roo Code uses a curated list of Baseten models. The default model is `zai-org/GLM-4.6`.
117+
118+
- **Multi-Cloud Capacity Management (MCM):** Baseten's multi-cloud infrastructure ensures high availability and low latency globally.
119+
120+
- **Support:** Baseten provides dedicated support for production deployments and can work with you on dedicated resources as you scale.
121+
122+
- **Pricing:** Current pricing is highly competitive and transparent. Prices typically range from $0.10-$6.00 per million tokens, making Baseten significantly more cost-effective than many closed-model alternatives while providing access to state-of-the-art open-source models.

‎sidebars.ts‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -161,6 +161,7 @@ const sidebars: SidebarsConfig = {
161161
label: 'Model Providers',
162162
items: [
163163
'providers/anthropic',
164+
'providers/baseten',
164165
'providers/claude-code',
165166
'providers/bedrock',
166167
'providers/cerebras',

0 commit comments

Comments
 (0)