Open-source models at scale. $5 free credits.
// At a glance
Free Tier
$5 credits · no card
// Free tier details
Available Models
LlamaMixtralQwenDeepSeek
Monthly Requests
$5 free credits
// Quick start
300">"text-purple-400">from openai 300">"text-purple-400">import OpenAI
client = OpenAI(
api_key=300">"YOUR_TOGETHER_KEY",
base_url=300">"https://api.together.xyz/v1",
)
response = client.chat.completions.create(
model=300">"meta-llama/Llama-3.3-70B-Instruct-Turbo",
messages=[{300">"role": 300">"user", 300">"content": 300">"Hello."}],
)
print(response.choices[0].message.content)
// Overview
Run Llama, Mixtral, and dozens of OSS models via a single API. Includes fine-tuning and dedicated GPU options.
// Pros
- Huge model catalog
- Fine-tuning support
- Generous $5 free credit
// Cons
- Credits expire in 3 months
- Some models slower than Groq
// Score breakdown
Reliability (35%) (from 2m ago health check)100/100
Free Tier Generosity (30%) (computed from quota, no-CC, no-phone fields)85/100
Documentation (20%) (human rating)90/100
Popularity (15%) (GitHub stars (log-normalised), or manual baseline)92/100
Methodology: apivault.dev/methodology
// Best for
OSS model hostingFine-tuningBatch inference
// Recent changes
May 18, 2026Added Llama 3.3 70B Turboadded