Ministral 3B API: Endpoint & Pricing
Ministral 3 3B 2512 is a 3-billion-parameter variant in Mistral’s Ministral 3 family, designed as a compact, efficient language model.
What is Ministral 3 3B 2512?
Ministral 3 3B 2512 is a small-scale, general-purpose language model developed by Mistral with around 3 billion parameters for efficient text understanding and generation. It is mainly used for lightweight conversational agents, code or content assistants, and applications that must run with limited compute or memory. It also suits experimentation, prototyping, and on-device or edge deployments where larger models are impractical. It belongs to Mistral’s Ministral 3 series of models, which comprise multiple sizes tuned for different performance and resource trade-offs.
Can you self-host Ministral 3 3B 2512?
No. We found no public weights for Ministral 3 3B 2512 on Hugging Face, so there is nothing to download, quantize or serve on your own GPUs: no VRAM budget, no vLLM or GGUF build to plan for. The only way to run it is over an API.
Skip the deploy — use LLM.API as your endpoint
No GPUs, no quantization trade-offs. Same model, OpenAI-compatible, up to 30% below list price.
Providers
Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Exacto (highest tool-calling accuracy).
List price by provider
InputOutputProvider list prices; the LLM.API discount applies on top.
| Provider | Input | Output | Cache read /M | Latency | Throughput | Uptime |
|---|---|---|---|---|---|---|
| in $0.1; out $0.1 per 1M tokens | 131K tokens | vision, tools, streaming, JSON, structured |
Prices and availability from the LLMAPI catalogue, updated nightly. Last updated 8 Oct 2026.
Try this model
Test Ministral 3 3B 2512 right here — free to start.
Suggestions for your first prompt
Code snippet
Call the model through the OpenAI-compatible API.
Switching from OpenAI? Change 2 lines.
- base_url="https://api.openai.com/v1"
- api_key="YOUR_OPENAI_KEY"
+ base_url="https://api.llmapi.ai/v1"
+ api_key="YOUR_LLMAPI_KEY"Everything else stays the same — SDK, message format, tools, streaming. Set model to ministral-3b-2512 and the call runs unchanged.
curl https://api.llmapi.ai/v1/chat/completions \
-H "Authorization: Bearer $LLMAPI_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "ministral-3b-2512",
"messages": [
{"role": "user", "content": "Give me three crisp launch checklist items."}
]
}'from openai import OpenAI
client = OpenAI(
api_key="YOUR_LLMAPI_KEY",
base_url="https://api.llmapi.ai/v1",
)
resp = client.chat.completions.create(
model="ministral-3b-2512",
messages=[
{"role": "system", "content": "You are a precise product assistant."},
{"role": "user", "content": "Give me three crisp launch checklist items."},
],
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.LLMAPI_KEY,
baseURL: "https://api.llmapi.ai/v1",
});
const resp = await client.chat.completions.create({
model: "ministral-3b-2512",
messages: [
{ role: "user", content: "Give me three crisp launch checklist items." },
],
});
console.log(resp.choices[0].message.content);stream = client.chat.completions.create(
model="ministral-3b-2512",
messages=[{"role": "user", "content": "Write a haiku about latency."}],
stream=True,
)
for chunk in stream:
delta = chunk.choices[0].delta.content
if delta:
print(delta, end="", flush=True){
"model": "ministral-3b-2512",
"messages": [
{"role": "system", "content": "You are a precise product assistant."},
{"role": "user", "content": "Give me three crisp launch checklist items."}
]
}ENDPOINTS
Supported endpoints & features
Base URL https://api.llmapi.ai/v1 · model ID ministral-3b-2512 · OpenAI-compatible.
POST /v1/chat/completionsChat Completions — supported
| Feature | Status | How to set it |
|---|---|---|
| Function calling (tools) | Supported | tools, tool_choice |
| Parallel tool calls | Not supported | parallel_tool_calls |
| JSON mode | Supported | response_format: {"type": "json_object"} |
| Structured outputs | Supported | response_format: {"type": "json_schema", …} |
| Streaming | Supported | stream: true |
| Vision (image input) | Supported | image_url content parts |
| Web search | Not supported | web_search |
ERRORS
Errors & fallback routing
| Code | What it means | What to do |
|---|---|---|
401 | Missing or invalid API key. | Send Authorization: Bearer YOUR_LLMAPI_KEY and check the key is active. |
429 | Rate limited, or the upstream provider is throttling the request. | Back off and retry with jitter; the gateway also retries the request on another provider where one is available. |
5xx | Upstream provider error or timeout. | Retry; fallback routing sends the retry to the next healthy provider for this model. |
When a provider fails or throttles, the request is routed to the next provider serving this model — the ones listed in the providers table above.
Ministral 3B in production
What the model is good at.
-
Conversational Chat
Handles multi-turn dialogue, follows instructions, and generates coherent, context-aware responses for general-purpose chat and assistance tasks.
-
Text Monitoring
Analyzes text to detect basic categories, topics, or potential issues for lightweight moderation, filtering, or routing scenarios.
-
Image Handling
Can be integrated into pipelines that associate text with images, enabling external systems to pair generated descriptions or prompts with visuals.
-
OCR Integration
Works with upstream OCR tools by interpreting extracted text, enabling summarization, classification, or transformation of document contents.
-
Text Translation
Supports multilingual text handling through translation-like tasks, enabling understanding and transformation of content between several major languages.
What you get when you call Ministral 3B through LLM.API.
-
Intelligent Model Routing
Dynamically route each request to the best model across providers using policies, evals, and metadata—no code changes when models, prices, or quotas shift.
One endpoint, every model -
Cost-Aware Optimization
Control spend with price-aware routing, per-project limits, and usage controls so you can experiment freely, cap risk, and hit performance targets within budget.
Max performance, min spend -
Automatic Resilient Fallbacks
Define provider- and model-level fallbacks so requests transparently retry on healthy models, keeping your app up during rate limits, outages, or provider regressions.
No-single-provider failure -
Full-Stack Observability
Trace every request across providers with logs, metrics, and structured events, so you can debug latency, errors, and quality issues from one unified view.
One pane of glass -
Task-Level Abstractions
Call high-level tasks—chat, embeddings, tools, rerank—from a single schema while LLM.API handles provider quirks, versioning, and feature differences under the hood.
Code to tasks, not vendors -
High-Throughput Batch
Submit massive batch jobs to any provider with automatic chunking, retries, and progress tracking so you can process millions of items reliably and cheaply.
Scale to millions safely
When to Use — When NOT to Use
Use it if...
- You need a very small, inexpensive model for large-scale batch text processing.
- Your use case involves lightweight classification, tagging, or routing on short inputs.
- You need fast experimentation with many parallel calls under tight cost constraints.
- Your use case involves simple prompt completion, short-form drafting, or boilerplate generation.
- You need an embedded model for edge or resource-constrained environments with limited memory.
- Your use case involves acting as a cheap first-pass filter before heavier models run.
Avoid if...
- You need state-of-the-art reasoning performance on complex, multi-step or ambiguous problems.
- Your workload requires high-quality long-form writing, such as reports or technical articles.
- You need strong coding assistance across multiple languages and complex software projects.
- Your workload requires handling very long context windows with consistent reasoning and recall.
- You need advanced multimodal capabilities like detailed image understanding or generation.
- Your workload requires top-tier safety, nuance, and domain expertise in sensitive applications.
COMMUNITY
What developers say about Ministral 3 3B 2512
Summarised from public developer discussions and write-ups. Opinions are the sources’, not LLM.API’s.
- Developers on benchmarking platforms report the model exhibits strong performance in general knowledge and ethics, though it struggles significantly with hallucinations and complex reasoning tasks compared to larger models.
- Technical reviews note that the model is highly efficient for edge deployments, supporting a large context window of up to 256K tokens while fitting within 8GB of VRAM in FP8.
- Developers appreciate the model's cost-efficiency for high-volume tasks like data extraction and support triage, citing its competitive pricing of $0.10 per million tokens.
SOURCES
Frequently Asked Questions
-
What is Ministral 3 3B 2512?
Ministral 3 3B 2512 is a 3B-parameter Mistral language model exposed through LLM.API for lightweight, low-cost text generation tasks.
-
What is Ministral 3 3B 2512 best suited for?
It is best for fast, inexpensive text tasks like drafting, rewriting, simple agents, and lightweight reasoning where latency and cost matter more than peak capability.
-
What modalities does Ministral 3 3B 2512 support on LLM.API?
On LLM.API, Ministral 3 3B 2512 is available as a text-only model, supporting prompt and completion in natural language.
-
What is the context window of Ministral 3 3B 2512?
Ministral 3 3B 2512 supports a 25,120-token context window, enabling relatively long conversations or documents in a single request.
-
How fast is Ministral 3 3B 2512 in terms of latency and throughput?
As a small 3B model, it typically delivers low first-token latency and high tokens-per-second throughput compared to larger Mistral models.
-
How is pricing for Ministral 3 3B 2512 handled on LLM.API?
Pricing is usage-based per 1,000 tokens, with exact input and output rates defined in the Ministral 3 3B 2512 section of LLM.API’s pricing page.
-
How do I call Ministral 3 3B 2512 through LLM.API?
Use the LLM.API chat or completion endpoint and set the model field to the Ministral 3 3B 2512 identifier documented in the LLM.API reference.
-
How does Ministral 3 3B 2512 compare to larger Mistral models?
It is cheaper and faster but generally less capable on complex reasoning, coding, and nuanced instruction-following than larger Mistral models.
-
Does Ministral 3 3B 2512 support tools or function calling via LLM.API?
If enabled by LLM.API, you can use the standard tools or function-calling schema with this model like any other supported chat model.
-
What are key limitations of Ministral 3 3B 2512?
It may struggle with very complex reasoning, domain-expert tasks, strict safety-sensitive use cases, and extremely long multi-step instructions despite its extended context.
COMPARE
Competitive Models
GLM-4 32B (0414-128k)
$0.1 in / $0.1 out per 1M tokens, 128,000 context — the closest current alternative to Ministral 3B.
Qwen3 VL 8B Instruct
$0.08 in / $0.5 out per 1M tokens, 131,072 context — the closest current alternative to Ministral 3B.
Llama 3.3 70B Instruct
$0.13 in / $0.4 out per 1M tokens, 131,072 context — the closest current alternative to Ministral 3B.
DeepSeek V3.2 is a large open-source Mixture-of-Experts language model from DeepSeek that emphasizes high reasoning performance and efficient long‑context inference. It is notable for its DeepSeek Sparse Attention and multi-latent attention mechanisms, which significantly cut compute and memory costs for long sequences.
Claude Opus 4.7
Up to 30%Claude Opus 4.7 is Anthropic’s most capable generally available large language model, designed for advanced coding, long-horizon agentic workflows, and high-resolution vision tasks. It emphasizes stronger multi-step reasoning, reliability on complex work, and improved instruction following compared to earlier Opus releases.
Kimi K2.6 (free)
Up to 30%Kimi K2.6 (free) is MoonshotAI’s open-source, multimodal Mixture-of-Experts model optimized for long-horizon coding, autonomous agents, and large-context reasoning. The free variant provides access to these capabilities via selected platforms and endpoints without direct model licensing costs.
Get one key to every model
Swap your API key. Keep your code.