Last refreshed 2026-11-05. Next refresh: weekly.
Why use Mistral Small 3 on Together AI?
Together AI offers Mistral Small 3 with pay-as-you-go pricing at $0.10/1M input tokens. Together AI is a platform for running open-source and proprietary LLMs with fast serverless and dedicated endpoints at competitive inference pricing.
Compare Mistral Small 3 across 3 providers to find the best fit for your use caseInput / 1M
$0.10
Output / 1M
$0.30
Cache
Not sourced
Batch
Not sourced
Setup recipe
Python + curlInstall
pip install togetherAuth
export TOGETHER_API_KEY=...Call
from together import Together
client = Together() # reads TOGETHER_API_KEY from env
response = client.chat.completions.create(
model="mistralai/Mistral-Small-24B-Instruct-2501",Model ID
mistralai/Mistral-Small-24B-Instruct-2501Request example
from together import Together
client = Together() # reads TOGETHER_API_KEY from env
response = client.chat.completions.create(
model="mistralai/Mistral-Small-24B-Instruct-2501",
messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)Gotchas
- Use provider model ID "mistralai/Mistral-Small-24B-Instruct-2501", not the LLMReference slug "mistral-small-3".
- Together uses "organization/model-name" format, e.g. "meta-llama/Llama-4-Scout-17B-16E-Instruct" or "Qwen/QwQ-32B". See the Together model catalog for the exact ID.
- The examples expect TOGETHER_API_KEY; rename it only if your application config maps the new variable.
Compare Mistral Small 3 Across Providers
| Provider | Input (per 1M) | Output (per 1M) |
|---|---|---|
| Together AI | $0.10 | $0.30 |
| OpenRouter | — | — |
| Vercel AI Gateway | $0.15 | $0.60 |
Pricing
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.10 |
| Output tokens | $0.30 |
Capabilities
JSON / Tool useStructured Outputs
About Mistral Small 3
Mistral Small 3 is a 24B-parameter, latency-optimized model that Mistral released on January 30, 2025, as both pretrained and instruction-tuned open-weight checkpoints. Mistral aims it at fast conversational assistants, low-latency function calling, fine-tuning for domain specialists, and local inference; once quantized, it can run on a single RTX 4090 or a MacBook with 32GB of RAM.
Get Started
Model Specs
Released2025-01-30
Parameters24B
Context32K
ArchitectureDecoder Only
Knowledge cutoff2023-10