Which is cheaper, Codestral-22B or Shieldstral 1.0 (3B)?
As of August 30, 2026, Shieldstral 1.0 (3B) is 83% cheaper on blended LLM API pricing. Shieldstral 1.0 (3B) costs $0.10 per million input tokens and $0.10 per million output tokens ($0.20 blended 1M-in + 1M-out). Codestral-22B costs $0.30 / $0.90 per million tokens ($1.20 blended). Sources: https://mistral.ai/technology/pricing and https://docs.mistral.ai/platform/pricing. The cheapest LLM for your app still depends on how many output tokens you generate.
How much does Codestral-22B vs Shieldstral 1.0 (3B) cost per million tokens?
Codestral-22B Mistral API pricing is $0.30 input and $0.90 output per million tokens via Mistral AI. Shieldstral 1.0 (3B) Mistral API pricing is $0.10 input and $0.10 output per million tokens via Mistral. Use those four numbers, not ChatGPT Plus or Claude Pro subscription prices, when you are comparing APIs.
Which is better for coding, Codestral-22B or Shieldstral 1.0 (3B)?
Codestral-22B has a published HumanEval score of 81.1%. Shieldstral 1.0 (3B) does not have that eval on this page yet.
What is the context window for Codestral-22B vs Shieldstral 1.0 (3B)?
Codestral-22B supports a 33K token context window, while Shieldstral 1.0 (3B) supports 33K tokens. Both models have the same context window size. Filling a larger window bills more input tokens, so the cheaper-per-million model can still cost more on long documents.
Which model performs better on benchmarks, Codestral-22B or Shieldstral 1.0 (3B)?
Shared benchmark scores for Codestral-22B and Shieldstral 1.0 (3B) are still limited. Check the comparison table for the evals each model has published. Check the comparison table for GPQA Diamond, SWE-bench, MMLU, HLE, and the other evals we track. Token price and benchmark score together are the usual LLM comparison, not either number alone.
Is Codestral-22B or Shieldstral 1.0 (3B) better for production use?
Both Codestral-22B and Shieldstral 1.0 (3B) are production API models. For cost-sensitive production traffic, Shieldstral 1.0 (3B) has better economics on the blended 1M-in + 1M-out scale. For maximum capability, weight the benchmark table for your domain. Many production stacks route cheap models for drafts and a frontier model for the hard turn, which is usually cheaper than sending everything to the expensive API.
Can I switch between Codestral-22B and Shieldstral 1.0 (3B) in my app?
Yes. Call Codestral-22B through Mistral AI and Shieldstral 1.0 (3B) through Mistral with separate API keys, or through a gateway that already wraps both. Keep prompts in tokens, not characters, when you estimate the invoice. AnotherWrapper templates can swap providers without rewriting auth and billing.
How accurate is this Codestral-22B vs Shieldstral 1.0 (3B) pricing data?
List rates are the latest published Mistral API pricing and Mistral API pricing, quoted per million tokens. Official rate sheets: https://mistral.ai/technology/pricing and https://docs.mistral.ai/platform/pricing. Prompt caching, batch APIs, and committed-use discounts can change the invoice. Benchmark scores are from official publications and independent evals, updated as new numbers land.
What is the output token limit for Codestral-22B and Shieldstral 1.0 (3B)?
Codestral-22B supports up to an unspecified number of output tokens per request, while Shieldstral 1.0 (3B) supports up to an unspecified number of output tokens. Output tokens are usually the expensive half of API pricing, so a higher max output cap is a capability, not a discount.
Which model has better throughput, Codestral-22B or Shieldstral 1.0 (3B)?
Neither Codestral-22B nor Shieldstral 1.0 (3B) has published throughput data on this page yet. Throughput (tokens per second) and time-to-first-token are why two models with the same per-million-token price can still feel different in chat and agents.
How do I switch the models in this Codestral-22B vs Shieldstral 1.0 (3B) comparison?
Use the model pickers at the top of this LLM comparison. Search any model in the index and the URL updates. Open Codestral-22B or Shieldstral 1.0 (3B) from the sidebar for the single-model API pricing page, or go back to the LLM pricing table to sort by cheapest blended cost.
