LimitDigest
Home / Compare / AI models / Mistral Large 4 vs Mistral Medium 3.5

Mistral Large 4 vs Mistral Medium 3.5

Official Mistral API prices and LMArena scores, checked 2026-10-31. Add up to 2 more models to compare 4.

Verdict

Mistral Large 4 is cheaper and wins more LMArena tasks.

Mistral Large 4 is 2.1x cheaper on input and 3.6x cheaper on output. Mistral Large 4 scores 10 points higher for agent tasks. Across 7 shared LMArena tasks, Mistral Large 4 leads 2 to 0 (5 tied). Mistral Large 4 has the longer context window (1M vs 256K tokens). A medium month (20M input + 5M output tokens) costs $24.50 on Mistral Large 4 and $67.50 on Mistral Medium 3.5.

  • Pick Mistral Large 4 if you want the lower bill ($24.50 vs $67.50 for a medium month), you care most about agent tasks (+10), you need the 1M context window or you reuse long prompts (cached input $0.07 vs $0.15).
  • Pick Mistral Medium 3.5 if nothing here clearly favors it (no lower bill, bigger context, cheaper cache or LMArena lead of 3+ points over Mistral Large 4).
Sample month

Medium: 20M input + 5M output tokens a month at list price (short-context rate, no caching or batch discounts).

Mistral Large 4vsMistral Medium 3.5
Price per 1M tokens
Input$0.70 (better)$1.50
Cached input$0.07 (better)$0.15
Output$2.10 (better)$7.50
Sample month, medium$24.50 (better)$67.50
Context
Context window1M (better)256K
Quality (LMArena relative score, out of 100)
Coding2 (better)0
Chat00
Creative writing00
Roleplay and characters00
Reasoning and math00
Vision00
Long documentsn/an/a
Cheap bulk and agents10 (better)0

Marked side is better: lower for prices, higher for context and scores. Bars are scaled to the larger of the two.

Spec
Mistral Large 4
Mistral Medium 3.5
Price per 1M tokens
Input$0.70$1.50
Cached input$0.07$0.15
Output$2.10$7.50
Blend 3:1$1.05$3
Sample month, medium$24.50$67.50
Context
Context window1M tokens256K tokens
Quality (LMArena, out of 100)
Coding2/100 #29 of 500/100 #48 of 50
Chat0/100 #43 of 510/100 #44 of 51
Creative writing0/100 #45 of 510/100 #42 of 51
Roleplay and characters0/100 #41 of 510/100 #43 of 51
Reasoning and math0/100 #26 of 510/100 #43 of 51
Vision0/100 #21 of 330/100 #33 of 33
Long documentsNot rankedNot ranked
Cheap bulk and agents10/100 #32 of 370/100 #37 of 37
Details
ProviderMistralMistral
API model idmistral-large-4mistral-medium-3.5
NotesNoneMistral deprecated Medium 3.5 on Oct 31, 2026 and points new integrations to Mistral Large 4. Last listed price shown.
Sources
Last checked2026-10-312026-10-30
Official pricingMistral pricingMistral pricing

API prices in USD per 1M tokens from each provider's official pricing page. Blend 3:1 = (3 x input + output) / 4. Sample month = stated input and output tokens at the list price, no caching. Quality: relative score out of 100 on each LMArena leaderboard (100 = top model we price on that board, n/a = not ranked there), data licensed CC BY 4.0. See how we pick.

Where each one ranks on LMArena

Relative score out of 100 on each leaderboard (100 = the top model we price there), with the rank among the models from our API price table that appear on that board.

  • Coding (Code Arena): Mistral Large 4 #29 of 50 (2/100), Mistral Medium 3.5 #48 of 50 (0/100)
  • Chat (Text Arena): Mistral Large 4 #43 of 51 (0/100), Mistral Medium 3.5 #44 of 51 (0/100)
  • Creative writing (Text Arena creative writing): Mistral Large 4 #45 of 51 (0/100), Mistral Medium 3.5 #42 of 51 (0/100)
  • Roleplay and characters (Text Arena multi-turn): Mistral Large 4 #41 of 51 (0/100), Mistral Medium 3.5 #43 of 51 (0/100)
  • Reasoning and math (Text Arena math): Mistral Large 4 #26 of 51 (0/100), Mistral Medium 3.5 #43 of 51 (0/100)
  • Vision (Vision Arena): Mistral Large 4 #21 of 33 (0/100), Mistral Medium 3.5 #33 of 33 (0/100)
  • Long documents (Document Arena): Mistral Large 4 not ranked, Mistral Medium 3.5 not ranked
  • Cheap bulk and agents (Agent Arena): Mistral Large 4 #32 of 37 (10/100), Mistral Medium 3.5 #37 of 37 (0/100)

Pricing details from the official pages

Mistral Large 4 lists $0.70 input, $2.10 output, $0.07 cached input per 1M tokens on Mistral pricing. Mistral Medium 3.5 lists $1.50 input, $7.50 output, $0.15 cached input per 1M tokens on Mistral pricing. Prices checked 2026-10-31.

  • Mistral Medium 3.5: Mistral deprecated Medium 3.5 on Oct 31, 2026 and points new integrations to Mistral Large 4. Last listed price shown.

Sample monthly bill at list price, no caching: Light (2M input + 0.5M output tokens): $2.45 vs $6.75. Medium (20M input + 5M output tokens): $24.50 vs $67.50. Heavy (200M input + 50M output tokens): $245 vs $675.

FAQ

Is Mistral Large 4 cheaper than Mistral Medium 3.5?

Mistral Large 4 costs $0.70 input and $2.10 output per 1M tokens; Mistral Medium 3.5 costs $1.50 and $7.50. For a medium month (20M input + 5M output tokens) Mistral Large 4 is cheaper: $24.50 vs $67.50.

Which is better for coding, Mistral Large 4 or Mistral Medium 3.5?

On LMArena's Code Arena Mistral Large 4 scores 2/100 (#29 of 50) and Mistral Medium 3.5 0/100 (#48 of 50), so Mistral Large 4 is ahead by 2 points.

Which has the bigger context window?

Mistral Large 4: 1M tokens. Mistral Medium 3.5: 256K tokens.

How much do Mistral Large 4 and Mistral Medium 3.5 cost per month?

Light (2M input + 0.5M output tokens): $2.45 vs $6.75. Medium (20M input + 5M output tokens): $24.50 vs $67.50. Heavy (200M input + 50M output tokens): $245 vs $675. Pick your own token counts in the plan vs API calculator.

Where do these numbers come from?

Prices come from each provider's official pricing page, checked daily. Quality is the relative score on the LMArena leaderboard dataset, licensed CC BY 4.0. Where a model is not ranked we show n/a.

Sources

Related