GPT-3.5-turbo vs Magistral Small.

model vs model · prices as of

Magistral Small takes 2 of 5 rounds against GPT-3.5-turbo. It is also the cheaper of the two to run, at $1.05 per million tokens against $2.73.

OpenAI

GPT-3.5-turbo

$2.73/M tokens

cheapest we can explain, at Poe · $0.355 per 1,000 pages

16k context

Poe
Mistral AIwins 2 rounds

Magistral Small

$1.05/M tokens

cheapest we can explain, at Eden AI · $0.126 per 1,000 pages

128k context · tool calling · open weights

Eden AI

List price in (lower wins)

$0.50/M

$0.50/M

List price out (lower wins)

$1.50/M

$1.50/M

Cheapest we can explain (lower wins)

$2.73/M at Poe

$1.05/M at Eden AI

Standard price (lower wins)

$3/M

$3/M

Cost per 1,000 pages (lower wins)

$0.355

$0.126

Context

16k tokens

128k tokens

Open weights

no

yes

Tool calling

no

yes

Structured output

no

not stated

Reads images

no

no

Reasoning

no

yes

Released

1 Mar 2023

17 Mar 2025

Providers selling it

18

6

Price we measure against

the maker's own

the maker's own

Trains on your prompts (no is better)

not stated

not stated

Providers selling both · 2

  • Eden AI$3 / $1.05
  • LLMTR$3 / $3

GPT-3.5-turbo then Magistral Small, per million tokens

Go further

Add a third or fourth model, or switch the weighting, in the tool.

What moved in the last seven days

  1. GPT-3.5-turboNous Portal$0.80 / $1.60 $1 / $225% dearer ·

Frequently asked

Which is cheaper, GPT-3.5-turbo or Magistral Small?
Magistral Small, at $1.05 per million tokens against $2.73 for GPT-3.5-turbo, three parts input to one part output, read on 16 Sept 2026.
Where is each one cheapest?
GPT-3.5-turbo: $2.73 at Poe. Magistral Small: $1.05 at Eden AI.
Can I buy both from one provider?
Yes. 2 providers sell both: Eden AI, LLMTR. The cheapest of them for the two together is Eden AI.
Which holds more context?
Magistral Small, at 128k tokens against 16k.

Prices exclude tax and are per million tokens, read 16 Sept 2026 from each provider's own catalog. The standard price blends three parts input to one part output; a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.

Related head-to-headsGPT-3.5-turbo vs Mistral Large 3GPT-3.5-turbo vs Qwen2.5 Coder 32B InstructGPT-3.5-turbo vs Qwen3 235B A22B Thinking 2507GPT-3.5-turbo vs Qwen3 30B A3B Thinking 2507GPT-3.5-turbo vs Qwen3 VL 30B A3B ThinkingMagistral Small vs Mistral Large 3Morecompare any two modelsevery model we track