Model Comparison

StarCoder2 15B vs Codestral 22B v0.1

Compare the listed model data for StarCoder2 15B by bigcode and Codestral 22B v0.1 by mistralai. Benchmark and pricing records are not a controlled test and may change.

Data last updated: 2026-10-16. Verify current model details with each provider.

What the listed data says

StarCoder2 15B: 16K tokens; Codestral 22B v0.1: 32K tokens. Codestral 22B v0.1 has the larger listed context window. StarCoder2 15B is listed for coding; Codestral 22B v0.1 is listed for coding, agentic workflows. Shared catalog benchmark entries: HumanEval: StarCoder2 15B 46.5%; Codestral 22B v0.1 81.1%; MultiPL-E: StarCoder2 15B 47.3%; Codestral 22B v0.1 73.9%; DS-1000: StarCoder2 15B 34.0%; Codestral 22B v0.1 53.3%; MBPP: StarCoder2 15B 65.2%; Codestral 22B v0.1 78.2%. These records are reference data, not a controlled head-to-head test; verify current pricing and specifications with each provider.

Catalog source links

Use these model and provider links to verify current specifications, pricing, and availability.

Computed differences from catalog data

  • The listed context windows are 16K tokens and 32K tokens; the larger is 2× the smaller.
  • On shared numeric benchmark entries, HumanEval: StarCoder2 15B − Codestral 22B v0.1 = -34.6 percentage points (46.5% vs 81.1%); MultiPL-E: StarCoder2 15B − Codestral 22B v0.1 = -26.6 percentage points (47.3% vs 73.9%); DS-1000: StarCoder2 15B − Codestral 22B v0.1 = -19.3 percentage points (34.0% vs 53.3%); MBPP: StarCoder2 15B − Codestral 22B v0.1 = -13 percentage points (65.2% vs 78.2%).
  • The first listed parameter quantities are 15.5B and 22B; the larger quantity is 1.42× the smaller. Total and active MoE parameters are not directly equivalent.
  • The catalog release dates are February 2024 and May 2024, about 3 months apart at the precision supplied.

Cost example: 1M input and 1M output tokens

A cost example for 1M input plus 1M output tokens cannot be calculated reliably because separate numeric input and output rates are not listed for both models.

Choose StarCoder2 15B if…

Choose StarCoder2 15B when your task aligns with catalog-listed uses or strengths such as IDE completion in rare or niche programming languages, Organizations prioritizing ethical AI data sourcing, Multi-language polyglot projects spanning 600+ languages, 600+ programming languages — broadest language coverage of any open code model.

Choose Codestral 22B v0.1 if…

Choose Codestral 22B v0.1 when your task aligns with catalog-listed uses or strengths such as Professional software development in enterprises, IDE code completion with FIM, Cross-language polyglot workflows, SQL and data science generation.

bigcodeB
StarCoder2 15B

Browse bigcode models

BigCode's StarCoder2 15B — trained on the Stack v2 with 600+ programming languages. Great for code completion and repository-level tasks.

Parameters: 15.5B

Context: 16K tokens

Catalog pricing: Apache 2.0 / BigCode OpenRAIL — free

bigcode Coding Open Source

View StarCoder2 15B model profile

Codestral 22B v0.1

Browse mistralai models

Mistral's flagship 22B code model — speaks 80+ languages, excellent at code generation, debugging, fill-in-the-middle, and tests.

Parameters: 22B

Context: 32K tokens

Catalog pricing: Non-Commercial free. Mistral API: €0.20/MTok input, €0.60/MTok output

mistralai Coding · Agentic

View Codestral 22B v0.1 model profile

Feature Comparison — StarCoder2 15B vs Codestral 22B v0.1

Catalog specifications for StarCoder2 15B and Codestral 22B v0.1
Specification StarCoder2 15B Codestral 22B v0.1
Provider bigcode mistralai
Parameters 15.5B 22B
Context window 16K tokens 32K tokens
Maximum output Not listed Not listed
Catalog pricing Apache 2.0 / BigCode OpenRAIL — free Non-Commercial free. Mistral API: €0.20/MTok input, €0.60/MTok output
Listed input/output cost per 1M tokens Separate numeric input/output rates not listed Separate numeric input/output rates not listed
License BigCode Open RAIL-M v1 Codestral Non-Commercial License
Coding support Listed Listed
Reasoning support Not listed Not listed
Vision input Not listed Not listed
Multimodal Not listed Not listed
Open weights Yes No
Benchmark: HumanEval 46.5% 81.1%
Benchmark: MultiPL-E 47.3% 73.9%
Benchmark: DS-1000 34.0% 53.3%
Benchmark: MBPP 65.2% 78.2%

Figures are transcribed from LITHOVEX model records. Benchmark versions and evaluation setups may differ; pricing and specifications can change. Check each provider's current documentation before making a decision.

Frequently Asked Questions

What is different between StarCoder2 15B and Codestral 22B v0.1?
StarCoder2 15B: 16K tokens; Codestral 22B v0.1: 32K tokens. Codestral 22B v0.1 has the larger listed context window. StarCoder2 15B is listed for coding; Codestral 22B v0.1 is listed for coding, agentic workflows. Shared catalog benchmark entries: HumanEval: StarCoder2 15B 46.5%; Codestral 22B v0.1 81.1%; MultiPL-E: StarCoder2 15B 47.3%; Codestral 22B v0.1 73.9%; DS-1000: StarCoder2 15B 34.0%; Codestral 22B v0.1 53.3%; MBPP: StarCoder2 15B 65.2%; Codestral 22B v0.1 78.2%. These records are reference data, not a controlled head-to-head test; verify current pricing and specifications with each provider.
What benchmark values are listed for StarCoder2 15B and Codestral 22B v0.1?
HumanEval: StarCoder2 15B 46.5%; Codestral 22B v0.1 81.1%; MultiPL-E: StarCoder2 15B 47.3%; Codestral 22B v0.1 73.9%; DS-1000: StarCoder2 15B 34.0%; Codestral 22B v0.1 53.3%; MBPP: StarCoder2 15B 65.2%; Codestral 22B v0.1 78.2%. The catalog values may use different versions and evaluation setups, so they are reference points rather than a controlled head-to-head result.
Which model has the larger listed context window?
StarCoder2 15B: 16K tokens; Codestral 22B v0.1: 32K tokens. Codestral 22B v0.1 has the larger listed context window.
What are the listed costs per million tokens for StarCoder2 15B and Codestral 22B v0.1?
StarCoder2 15B: Separate numeric input/output rates not listed (Apache 2.0 / BigCode OpenRAIL — free). Codestral 22B v0.1: Separate numeric input/output rates not listed (Non-Commercial free. Mistral API: €0.20/MTok input, €0.60/MTok output). Where separate numeric input and output rates are available, the worked example uses 1M input plus 1M output tokens. Confirm current terms with each provider.
How far apart are the listed release dates for StarCoder2 15B and Codestral 22B v0.1?
The catalog lists StarCoder2 15B as February 2024 and Codestral 22B v0.1 as May 2024, about 3 months apart at the precision supplied.
When was the comparison data last updated?
The model catalog records 2026-10-16 as its data update date. Provider specifications and prices may have changed since then.
How should I choose between StarCoder2 15B and Codestral 22B v0.1?
Compare the listed specifications and capabilities, review the benchmark notes, then test both with the same representative prompt for your task. This page does not claim a universal winner.

Explore these models and related comparisons

Read the StarCoder2 15B model profile and Codestral 22B v0.1 model profile, or browse the model directory.

Return to the comparison directory

Still deciding? Try both free — no commitment.

Create your free LITHOVEX account and compare StarCoder2 15B and Codestral 22B v0.1 on your actual tasks. No credit card, no API key, no subscriptions needed.

Try StarCoder2 15B Try Codestral 22B v0.1