Mistral Medium 3 logoMistral Medium 3
vs
Mistral Medium 3.5 logoMistral Medium 3.5

Mistral Medium 3 vs Mistral Medium 3.5: Which is Better in 2026?

A comprehensive comparison of Mistral Medium 3 and Mistral Medium 3.5 covering features, pricing, use cases, and which tool is the right choice for your needs.

⚡ Quick Verdict

Choose Mistral Medium 3 if:

  • You want more affordable paid plans (from $0.4/mo)
  • You need 74.5% on mmlu-pro — competitive with gpt-4o and claude sonnet 3.7 at launch or 128k context window — suitable for long-document analysis and large codebase ingestion

Choose Mistral Medium 3.5 if:

  • You want a free tier to get started without commitment
  • You need 128b dense model (merged: instruction-following + reasoning + coding) or 256k token context window

Mistral Medium 3 vs Mistral Medium 3.5: At a Glance

Attribute
Mistral Medium 3
Mistral Medium 3.5
Pricing Model
Paid
Freemium
Starting Price
Starting at $0.40/month
Starting at $1.5/month
Free Tier
✗ No
✓ Yes
Category
llm-apis
llm-apis
Features Count
9 features
9 features
Shared Features
0 features in common

Pricing Comparison: Mistral Medium 3 vs Mistral Medium 3.5

Understanding the pricing differences between Mistral Medium 3 and Mistral Medium 3.5 is crucial for making the right choice. Here's how their plans compare side by side.

Mistral Medium 3 Pricing

Starter$0.40/month
Starter$2.00/month
View full Mistral Medium 3 pricing →

Mistral Medium 3.5 Pricing

API:$1.5/month
Starter$7.5/month
View full Mistral Medium 3.5 pricing →

💡 Pricing takeaway: Mistral Medium 3.5 has an edge with a free tier, letting you start without commitment. Compare the specific plans to find the best value for your use case.

Feature-by-Feature Comparison

Here's how every feature from Mistral Medium 3 and Mistral Medium 3.5 stacks up.

Feature
Mistral Medium 3
Mistral Medium 3.5
74.5% on MMLU-Pro — competitive with GPT-4o and Claude Sonnet 3.7 at launch
128K context window — suitable for long-document analysis and large codebase ingestion
Multimodal vision via Pixtral architecture — image understanding built in, no separate model required
40+ language support — strong multilingual coverage for global enterprise use cases
Strong coding performance: top-tier on HumanEval and LiveCodeBench at mid-tier pricing
Available on major cloud marketplaces: AWS Bedrock, Azure AI Foundry, Google Vertex AI
Tool use and function calling — ready for agentic workflows and API integrations
Fine-tuning available for enterprise customers via Mistral La Plateforme
Priced 5–10× cheaper than GPT-4o and Claude Sonnet at equivalent capability tier
128B dense model (merged: instruction-following + reasoning + coding)
256k token context window
77.6% on SWE-Bench Verified (beats Devstral 2 and Qwen3.5 397B A17B)
91.4 on τ³-Telecom (strong agentic capabilities)
Configurable reasoning effort per request
Vision encoder trained from scratch — handles variable image sizes and aspect ratios
Open weights under modified MIT license (self-hostable on 4 GPUs)
Powers Mistral Vibe remote coding agents and Le Chat Work mode
Async cloud coding sessions with GitHub, Linear, Jira, Sentry integrations

What Makes Each Tool Unique

🔵 Unique to Mistral Medium 3

Features available in Mistral Medium 3 but not in Mistral Medium 3.5:

  • 74.5% on MMLU-Pro — competitive with GPT-4o and Claude Sonnet 3.7 at launch
  • 128K context window — suitable for long-document analysis and large codebase ingestion
  • Multimodal vision via Pixtral architecture — image understanding built in, no separate model required
  • 40+ language support — strong multilingual coverage for global enterprise use cases
  • Strong coding performance: top-tier on HumanEval and LiveCodeBench at mid-tier pricing
  • Available on major cloud marketplaces: AWS Bedrock, Azure AI Foundry, Google Vertex AI
  • Tool use and function calling — ready for agentic workflows and API integrations
  • Fine-tuning available for enterprise customers via Mistral La Plateforme
  • Priced 5–10× cheaper than GPT-4o and Claude Sonnet at equivalent capability tier

🟣 Unique to Mistral Medium 3.5

Features available in Mistral Medium 3.5 but not in Mistral Medium 3:

  • 128B dense model (merged: instruction-following + reasoning + coding)
  • 256k token context window
  • 77.6% on SWE-Bench Verified (beats Devstral 2 and Qwen3.5 397B A17B)
  • 91.4 on τ³-Telecom (strong agentic capabilities)
  • Configurable reasoning effort per request
  • Vision encoder trained from scratch — handles variable image sizes and aspect ratios
  • Open weights under modified MIT license (self-hostable on 4 GPUs)
  • Powers Mistral Vibe remote coding agents and Le Chat Work mode
  • Async cloud coding sessions with GitHub, Linear, Jira, Sentry integrations

Use Case Recommendations

Best for: Mistral Medium 3

Mistral AI's enterprise-grade mid-tier model released May 7, 2025. Positioned as 'medium is the new large' — delivers GPT-4o and Claude Sonnet 3.7-level performance at a fraction of the cost: $0.40/M input, $2.00/M output. 128K context window, multimodal (vision via Pixtral architecture), and strong on coding (MMLU-Pro 74.5%), STEM reasoning, and multilingual tasks across 40+ languages. Available via Mistral La Plateforme API, Amazon Bedrock, Azure AI, and Google Cloud Vertex AI.

Ideal use cases:

  • Teams or individuals who need 74.5% on mmlu-pro — competitive with gpt-4o and claude sonnet 3.7 at launch
  • Teams or individuals who need 128k context window — suitable for long-document analysis and large codebase ingestion
  • Teams or individuals who need multimodal vision via pixtral architecture — image understanding built in, no separate model required
  • Teams or individuals who need 40+ language support — strong multilingual coverage for global enterprise use cases
  • Anyone focused on mistral workflows
  • Anyone focused on llm workflows
Try Mistral Medium 3

Best for: Mistral Medium 3.5

Mistral's first flagship merged model, released May 22, 2026. A dense 128B model with a 256k context window that handles instruction-following, reasoning, and coding in a single set of weights. Available as open weights (modified MIT license) and powers Mistral Vibe remote coding agents and Le Chat's new Work mode. SWE-Bench Verified: 77.6%. API: $1.5/M input, $7.5/M output.

Ideal use cases:

  • Teams or individuals who need 128b dense model (merged: instruction-following + reasoning + coding)
  • Teams or individuals who need 256k token context window
  • Teams or individuals who need 77.6% on swe-bench verified (beats devstral 2 and qwen3.5 397b a17b)
  • Teams or individuals who need 91.4 on τ³-telecom (strong agentic capabilities)
  • Anyone focused on mistral workflows
  • Anyone focused on llm workflows
Try Mistral Medium 3.5

🔧 Other llm-apis Tools to Consider

Mistral Medium 3 and Mistral Medium 3.5 aren't the only options. Here are other popular tools in the same space:

Frequently Asked Questions

Is Mistral Medium 3 better than Mistral Medium 3.5?

It depends on your needs. Mistral Medium 3 offers 9 key features including 74.5% on MMLU-Pro — competitive with GPT-4o and Claude Sonnet 3.7 at launch and 128K context window — suitable for long-document analysis and large codebase ingestion, while Mistral Medium 3.5 provides 9 features including 128B dense model (merged: instruction-following + reasoning + coding) and 256k token context window. Mistral Medium 3 uses a paid model, while Mistral Medium 3.5 is freemium with free access available. Choose based on which features and pricing model align with your requirements.

Is Mistral Medium 3 cheaper than Mistral Medium 3.5?

Mistral Medium 3 is cheaper, starting at $0.40/month compared to Mistral Medium 3.5's $1.5/month. Mistral Medium 3.5 offers a free tier, making it easier to get started. Always check the official websites for the most current pricing.

Can I use Mistral Medium 3 and Mistral Medium 3.5 together?

Yes, many users combine Mistral Medium 3 and Mistral Medium 3.5 in their workflow. Mistral Medium 3 excels at 74.5% on mmlu-pro — competitive with gpt-4o and claude sonnet 3.7 at launch, while Mistral Medium 3.5 shines with 128b dense model (merged: instruction-following + reasoning + coding). Using both allows you to leverage the strengths of each tool, though this means managing two subscriptions — though free tiers can help manage costs.

What's the main difference between Mistral Medium 3 and Mistral Medium 3.5?

While both are llm-apis tools, Mistral Medium 3 emphasizes 74.5% on mmlu-pro — competitive with gpt-4o and claude sonnet 3.7 at launch, whereas Mistral Medium 3.5 is known for 128b dense model (merged: instruction-following + reasoning + coding). The best choice depends on your specific workflow and feature priorities.

Learn More

Related Comparisons