Mistral Small 2603
Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.
mistral-small-2603Released 16 Mar 2026 (Mar 2026)Pricing
List prices at Mistral AI, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $0.15 | $0.60 | |
| Cached input 90% off input | $0.015 | 90% off input |
What it costs a month
At Mistral AI list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$77.85
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$61.20
- Coding agent1,000 requests a day, 40K in / 2K out, cached$103
- Overnight batch20,000 requests a day, 4K in / 0.4K out$504
Where to run it
1 host serve Mistral Small 2603. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
Mistral AILab API mistral-small-2603 | $0.15 / $0.60 | mistral-small-2603 |
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- TextImages
- Produces
- Text
- Effort settings the source lists
- nonehigh
- Features
- Assistant prefill
- Open weights
- License: Apache-2.0Runs in about 286.6 GB of GPU memory
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than Mistral AI.
Smaller Mistral AI models
- 10% less
- 43% less
Other frontier labs
- 24% less
- 24% less
Open-weight picks
- 4% less
- 32% less
- 39% less
- 46% less
Model line
Newer and older versions, at each lab's own price.
- Mar 2026Mistral Small 2603Latest$0.15 / $0.60
- UndatedMistral Small 2503$0.10 / $0.30
- UndatedMistral Small 2402$1.00 / $3.00
- UndatedMistral Small 4$0.15 / $0.60
Call it
Using Mistral AI's API.
from litellm import completion
response = completion(
model="mistral/mistral-small-2603",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does Mistral Small 2603 cost?
On Mistral AI, Mistral Small 2603 costs $0.15 per million input tokens and $0.60 per million output tokens. Cached input costs $0.015 per million tokens.
Which APIs offer Mistral Small 2603?
Mistral Small 2603 runs on 1 API: Mistral AI.
What is Mistral Small 2603's context window?
262K tokens, with up to 262K tokens in one reply.
Is Mistral Small 2603 being retired?
No retirement date is scheduled for it on any API we track.
Is there a newer version of Mistral Small 2603?
No. Mistral Small 2603 is the newest model in its line.
What are cheaper alternatives to Mistral Small 2603?
Models that cost less and keep most of what it can do: Zai Glm 5.3 Flash on Mistral AI (10% less), Claude Haiku 5.5 on Anthropic (24% less), GPT-6 Luna on OpenAI (24% less) and Qwen3.8 27B on EmpirioLabs AI (4% less). Savings are on a mix of 3 input tokens to 1 output token.
When was Mistral Small 2603 released?
Mistral AI released it on 16 Mar 2026.
What is Mistral Small 2603's API model id?
Use mistral-small-2603 with Mistral AI.
Sources
Raw catalog entry for mistral-small-2603 on Mistral AI
{
"cache_read_input_token_cost": 1.5e-8,
"input_cost_per_token": 1.5e-7,
"litellm_provider": "mistral",
"max_input_tokens": 262144,
"max_output_tokens": 262144,
"mode": "chat",
"output_cost_per_token": 6e-7,
"reasoning_effort_levels": [
"none",
"high"
],
"source": "https://docs.mistral.ai/models/model-cards/mistral-small-4-0-26-03",
"supports_assistant_prefill": true,
"supports_function_calling": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_vision": true
}