Skip to content
ModelCap

Head-to-head comparison

Gemma 3 4B vs Codestral 2508

Gemma 3 4B (Google) and Codestral 2508 (Mistral AI) compared on the ModelCap Index, API price, context window, provider availability, weight access and every public benchmark board they share. Figures are the same ones shown on the live rankings; nothing here is a hidden score.

Snapshot as of 18 August 2026

As of 18 August 2026, Gemma 3 4B holds the stronger ModelCap Index position (#195 vs #227); Gemma 3 4B is 9× cheaper per output token ($0.10/1M vs $0.90/1M); and Codestral 2508 offers the longer context window (256K vs 131K tokens).

Which should you choose?

Choose Gemma 3 4B if…

  • you want the stronger overall ModelCap Index position — #195 against #227 (10.9 vs 0.0 points).
  • API cost matters — $0.10/1M output tokens against $0.90/1M, about 9× cheaper.
  • your workload is prompt-heavy — input tokens cost $0.05/1M against $0.30/1M.

Choose Codestral 2508 if…

  • you need the longer context window — 256K tokens against 131K.
  • you want the more recently listed model — Codestral 2508 was listed 1 August 2025, Gemma 3 4B 13 March 2025.

Gemma 3 4B vs Codestral 2508: specs, pricing and context

Specification comparison of Gemma 3 4B and Codestral 2508
FieldGemma 3 4BGoogleCodestral 2508Mistral AI
ModelCap Index position#195#227
Index score10.90.0
EvidenceMeasuredEstimated
Input price / 1M tokens$0.05$0.30
Output price / 1M tokens$0.10$0.90
Context window131K tokens256K tokens
Max output tokens16K—
API providers listed11
Weight accessGated accessAPI only
Input modalitiesText, ImageText, File
First listed13 March 20251 August 2025
PublisherGoogleMistral AI

Evidence: 3 public benchmark observations across 3 boards · publisher corpus prior · leave-one-anchor-out calibrated. Prices are the lowest listed API offer per million tokens observed on the OpenRouter catalogue.

Benchmark scores: Gemma 3 4B vs Codestral 2508

Public benchmark boards where Gemma 3 4B or Codestral 2508 has a published result
BoardGemma 3 4BCodestral 2508Leads
Arena codingArena (LMArena)12741250–1298—Only one result
BFCL V4UC Berkeley (Gorilla)19.6Prompt—Only one result
Artificial Analysis Intelligence IndexArtificial Analysis1.0intelligence-index—Only one result

Scores are the sources' own published figures for each model's best evaluated configuration; ModelCap never re-runs a benchmark.

Want a different pairing? Open the interactive comparison tool to swap either model for any current ranked language model.

Gemma 3 4B vs Codestral 2508: common questions

Is Gemma 3 4B better than Codestral 2508?

Gemma 3 4B ranks higher on the ModelCap Index as of 18 August 2026: #195 against #227. That is a capability ranking built from public benchmark evidence with published uncertainty; whether it is "better" for you also depends on price, context and where you can run it.

Is Gemma 3 4B cheaper than Codestral 2508?

Gemma 3 4B is cheaper on output tokens: $0.10/1M against $0.90/1M. Input tokens are $0.05/1M for Gemma 3 4B and $0.30/1M for Codestral 2508. Prices are the lowest listed API offer ModelCap observed, in USD per million tokens.

Which has the bigger context window, Gemma 3 4B or Codestral 2508?

Codestral 2508 has the larger context window: 256K tokens against 131K.

Which is better for coding, Gemma 3 4B or Codestral 2508?

The two models do not share a coding benchmark board on ModelCap yet, so no head-to-head coding score is published; the ModelCap Index position is the closest overall signal.

Are Gemma 3 4B and Codestral 2508 open-weight models?

Gemma 3 4B: Gated access. Codestral 2508: API only. Open weights mean the checkpoint can be downloaded and self-hosted under its licence; API-only models are available solely through hosted endpoints.

Where can I run Gemma 3 4B and Codestral 2508?

ModelCap currently lists 1 API provider for Gemma 3 4B and 1 for Codestral 2508, from the OpenRouter catalogue snapshot the site serves; each model page lists the providers and their prices.

How current is this Gemma 3 4B vs Codestral 2508 comparison?

Every figure comes from the sealed ModelCap dataset published 18 August 2026; the page re-renders within a minute of each data refresh, and the ModelCap Index positions are the same ones shown on the live rankings.

Explore further