SkillBoss Unified AI Gateway
One API key. 100+ models. Zero vendor accounts.
Why Use a Gateway?
| Problem |
SkillBoss Solution |
| Multiple API keys |
One key for everything |
| Multiple bills |
Single unified billing |
| Rate limits |
Auto-failover to alternatives |
| Vendor lock-in |
Switch models with one param |
Models Included
Chat/Reasoning
- Claude: 4.5 Opus, 4.5 Sonnet, 3.7 Sonnet, Haiku
- GPT: GPT-5, GPT-4.1, o4-mini, o3-mini
- Gemini: 2.5 Pro, 2.5 Flash
- DeepSeek: R1, V3
- Llama: 3.1 405B, 70B
- Mistral: Large
- Cohere: Command R+
Image Generation
- DALL-E 3, Flux Schnell/Dev, Minimax
Video Generation
- Veo 3.1, Minimax Video-01
Audio
- Whisper (STT), ElevenLabs (TTS), Minimax TTS
Data & Tools
- Perplexity Search, Firecrawl, Reducto, more
Usage
# Just change the model parameter
curl https://api.heybossai.com/v1/chat/completions \
-H "Authorization: Bearer $SKILLBOSS_API_KEY" \
-d '{
"model": "claude-4-5-sonnet", # or gpt-5, gemini-2.5-pro, etc.
"messages": [{"role": "user", "content": "Hello!"}]
}'
OpenAI SDK Compatible
from openai import OpenAI
client = OpenAI(
api_key="sk-your-skillboss-key",
base_url="https://api.heybossai.com/v1"
)
# Use any model
response = client.chat.completions.create(
model="claude-4-5-sonnet", # or any other model
messages=[{"role": "user", "content": "Hello!"}]
)
Features
- Auto-failover - If Claude is down, fallback to GPT
- Streaming - Real-time responses
- Function calling - Tool use supported
- Vision - Image inputs for supported models
- Caching - Prompt caching where available
Pricing
Pass-through pricing (same as vendor) + no markup.
Get started: https://skillboss.co/console
1---2name: skillboss-gateway3description: Access 100+ AI models with one API key. OpenAI-compatible. Auto-failover, unified billing, no vendor accounts needed.4license: MIT5---6
7# SkillBoss Unified AI Gateway
8
9**One API key. 100+ models. Zero vendor accounts.**
10
11## Why Use a Gateway?
12
13| Problem | SkillBoss Solution |
14|---------|-------------------|
15| Multiple API keys | One key for everything |
16| Multiple bills | Single unified billing |
17| Rate limits | Auto-failover to alternatives |
18| Vendor lock-in | Switch models with one param |
19
20## Models Included
21
22### Chat/Reasoning
23- **Claude**: 4.5 Opus, 4.5 Sonnet, 3.7 Sonnet, Haiku
24- **GPT**: GPT-5, GPT-4.1, o4-mini, o3-mini
25- **Gemini**: 2.5 Pro, 2.5 Flash
26- **DeepSeek**: R1, V3
27- **Llama**: 3.1 405B, 70B
28- **Mistral**: Large
29- **Cohere**: Command R+
30
31### Image Generation
32- DALL-E 3, Flux Schnell/Dev, Minimax
33
34### Video Generation
35- Veo 3.1, Minimax Video-01
36
37### Audio
38- Whisper (STT), ElevenLabs (TTS), Minimax TTS
39
40### Data & Tools
41- Perplexity Search, Firecrawl, Reducto, more
42
43## Usage
44
45```bash
46# Just change the model parameter
47curl https://api.heybossai.com/v1/chat/completions \
48 -H "Authorization: Bearer $SKILLBOSS_API_KEY" \
49 -d '{
50 "model": "claude-4-5-sonnet", # or gpt-5, gemini-2.5-pro, etc.
51 "messages": [{"role": "user", "content": "Hello!"}]
52 }'
53```
54
55## OpenAI SDK Compatible
56
57```python
58from openai import OpenAI
59
60client = OpenAI(
61 api_key="sk-your-skillboss-key",
62 base_url="https://api.heybossai.com/v1"
63)
64
65# Use any model
66response = client.chat.completions.create(
67 model="claude-4-5-sonnet", # or any other model
68 messages=[{"role": "user", "content": "Hello!"}]
69)
70```
71
72## Features
73
74- **Auto-failover** - If Claude is down, fallback to GPT
75- **Streaming** - Real-time responses
76- **Function calling** - Tool use supported
77- **Vision** - Image inputs for supported models
78- **Caching** - Prompt caching where available
79
80## Pricing
81
82Pass-through pricing (same as vendor) + no markup.
83
84Get started: https://skillboss.co/console
modbender