Model comparison

Mistral Medium 3.5 vs Sonar

Mistral Medium 3.5 is the stronger model overall, scoring 40.2 to 38.5 on the Noometry Index. Sonar costs 3.0× less per token, which makes it the better buy when Mistral Medium 3.5's lead doesn't matter for your workload.

Last verified October 11, 2026. 0 shared benchmarks.

Mistral Medium 3.5

Mistral AI

40.2

Rank #152 Confirmed

Sonar

Perplexity

38.5

Rank #187 Confirmed

Summary

The widest gap is in writing & preference, where Mistral Medium 3.5 leads 58.5 to 52.6.
Sonar is cheaper at $1 / $1 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium 3.5.
Mistral Medium 3.5 accepts more context: 262K tokens versus 128K.
Mistral Medium 3.5 has downloadable open weights; the other is API-only.

Side by side

Mistral Medium 3.5 and Sonar specifications
	Mistral Medium 3.5	Sonar
Provider	Mistral AI	Perplexity
Noometry Index	40.2	38.5
Released	—	2024-01-01
Weights	Open	Proprietary
Context window	262K	128K
Max output	210K	4K
Input $ / M tokens	$1.50	$1
Output $ / M tokens	$7.50	$1
Results tracked	22	7

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Too close to call

Mistral Medium 3.5: 36.0 (#213), Sonar: 35.7 (#221)

Coding benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LMArena WebDev	1264	—
LiveBench Coding	—	35.1%
LMArena Coding	1461	—

Reasoning Sonar leads

Mistral Medium 3.5: 17.3 (#295), Sonar: 21.1 (#227)

Reasoning benchmarks
Benchmark	Mistral Medium 3.5	Sonar
Kagi LLM Benchmark	41.4%	—
NYT Connections (extended)	12.9%	—
LiveBench Reasoning	—	46.3%
LMArena Hard Prompts	1436	—
LiveBench Data Analysis	—	37.9%
Epoch Capabilities Index	141.35	—
LiveBench	—	46.9%

Math Mistral Medium 3.5 leads

Mistral Medium 3.5: 39.1 (#113), Sonar: 33.7 (#200)

Math benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LiveBench Math	—	41.6%
LMArena Math	1431	—

Knowledge Not comparable

Mistral Medium 3.5: 40.0 (#126), Sonar: —

Knowledge benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LMArena Expert	1432	—

Multimodal Not comparable

Mistral Medium 3.5: 38.3 (#65), Sonar: —

Multimodal benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LMArena Vision	1223	—

Multilingual Not comparable

Mistral Medium 3.5: 51.9 (#100), Sonar: —

Multilingual benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LMArena Non-English	1404	—
LMArena Chinese	1442	—
LMArena French	1448	—
LMArena German	1451	—
LMArena Korean	1385	—
LMArena Russian	1395	—
LMArena Spanish	1409	—

Instruction Following Mistral Medium 3.5 leads

Mistral Medium 3.5: 74.6 (#90), Sonar: 71.4 (#150)

Instruction Following benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LiveBench Instruction Following	—	76.2%
LMArena Instruction Following	1415	—

Long Context Not comparable

Mistral Medium 3.5: 43.2 (#103), Sonar: —

Long Context benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LMArena Longer Query	1415	—

Writing & Preference Mistral Medium 3.5 leads

Mistral Medium 3.5: 58.5 (#117), Sonar: 52.6 (#167)

Writing & Preference benchmarks
Benchmark	Mistral Medium 3.5	Sonar
LMArena Text	1421	—
LMArena Creative Writing	1374	—
EQ-Bench 4	993	—
LMArena Multi-Turn	1423	—
LiveBench Language	—	44.1%

Frequently asked questions

Is Mistral Medium 3.5 better than Sonar?

Which is cheaper, Mistral Medium 3.5 or Sonar?

Sonar is cheaper. It lists at $1 per million input tokens and $1 per million output tokens; Mistral Medium 3.5 lists at $1.50 and $7.50.

Is Mistral Medium 3.5 or Sonar better for coding?

They score almost the same on coding (36.0 vs 35.7); test both on your own repository before choosing.

Which has the bigger context window?

Mistral Medium 3.5 does, with 262K tokens against 128K.

How many benchmarks do Mistral Medium 3.5 and Sonar share?

0 benchmarks have published results for both models. Mistral Medium 3.5 has 22 scored results on Noometry and Sonar has 7.