Noometry

Mistral AI, open weights

Mistral 7B

Mistral 7B by Mistral AI ranks 351st of 354 ranked models on the Noometry Index as of October 2026, with a score of 23.0. Its strongest category is long context, where it ranks 271st. API pricing starts at $0.25 per million input tokens and $0.25 per million output tokens, with a 8K-token context window.

Last verified

Noometry rank
#351 of 354
Index score
23.0
Evidence
Confirmed 37 results
Provider
Mistral AI
Released
September 27, 2023
Weights
Open weights
Reasoning
No
Context window
8K
Max output
8K
Input price
$0.25 / M
Output price
$0.25 / M
Blended price
$0.25 / M
Output speed
Not measured
Value
#67 of 219
Knowledge cutoff
December 2023
Input
text

Category scores

Each category score combines every public result we have in that category.

  1. Coding 26.4
  2. Reasoning 13.1
  3. Math 8.1
  4. Knowledge 7.4
  5. Multilingual 25.8
  6. Instruction Following 54.2
  7. Long Context 32.2
  8. Writing & Preference 30.7
CategoryScoreRankResults
Coding26.4#3263
Reasoning13.1#3363
Math8.1#3253
Knowledge7.4#3112
Multilingual25.8#2831
Instruction Following54.2#2801
Long Context32.2#2711
Writing & Preference30.7#2863

Strengths and weaknesses

Categories where Mistral 7B places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Long Context32.2−8.7#271 of 296, top 92%
Writing & Preference30.7−23.1#286 of 312, top 92%
Instruction Following54.2−17.1#280 of 305, top 92%

Weakest categories

CategoryScorevs medianRank
Math8.1−28.5#325 of 327, top 100%
Knowledge7.4−30.0#311 of 314, top 100%
Reasoning13.1−10.5#336 of 350, top 96%

Closest competitors

The models ranked just above and below Mistral 7B. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Claude 2#34625.0——Compare
DeepSeek LLM 67B#34724.9——Compare
Llama 13b#34824.4——Compare
Llama 2-70B#34924.4——Compare
GPT-3.5-turbo#35023.2$0.75—Compare
Llama 3.1-8B#35223.0$0.0575—Compare
Gemma 3 1B#35321.1——Compare
Llama 3.2 1B#35420.1$0.0705—Compare

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

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
BigCodeBench Instruct19.5%#62 of 64, top 97%BigCodeBench2024-05-22
LMArena Coding1082#276 of 294, top 94%LMArena2026-10-08
LMArena Coding1018LMArena2026-10-08
BigCodeBench Complete23.5%BigCodeBench2024-05-22
BigCodeBench Complete27.3%#63 of 66, top 96%BigCodeBench2024-05-22
HumanEval+36%#39 of 45, top 87%EvalPlus
HumanEval+23.8%EvalPlus
MBPP+37%EvalPlus
MBPP+42.1%#36 of 38, top 95%EvalPlus

Reasoning

BenchmarkScorePositionSettingSourceDate
Chess Puzzles0%#123 of 129, top 96%Epoch AI2026-08-30
LMArena Hard Prompts1002LMArena2026-10-08
LMArena Hard Prompts1067#278 of 297, top 94%LMArena2026-10-08
DTBench42.5%#149 of 151, top 99%Epoch AI
Adversarial NLI47.1%#8 of 9, top 89%Epoch AI
BIG-Bench Hard56.1%#16 of 27, top 60%Epoch AI
Epoch Capabilities Index112.21#188 of 213, top 89%Epoch AI2023-09-27
Epoch Capabilities Index108.99Epoch AI2024-05-27
HellaSwag81%#16 of 29, top 56%Epoch AI
PIQA83%#12 of 27, top 45%Epoch AI
PIQA82.2%Epoch AI
PIQA82.2%Epoch AI
WinoGrande75.3%#23 of 43, top 54%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
OTIS Mock AIME 2024-20250.3%#172 of 173, top 100%Epoch AI2026-08-30
LMArena Math1027LMArena2026-10-08
LMArena Math1085#270 of 285, top 95%LMArena2026-10-08
MATH Level 53.7%#78 of 79, top 99%Epoch AI2025-01-27
MATH Level 53.6%Epoch AI2025-01-27
GSM8K54.4%#20 of 38, top 53%Epoch AI
GSM8K35.4%Epoch AI

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond15.2%#185 of 186, top 100%Epoch AI2025-01-27
GPQA Diamond13.2%Epoch AI2025-01-27
LMArena Expert954LMArena2026-10-08
LMArena Expert1036#267 of 273, top 98%LMArena2026-10-08
ARC (AI2) Challenge78.6%#13 of 39, top 34%Epoch AI
BoolQ83.2%Epoch AI
BoolQ87.4%#4 of 23, top 18%Epoch AI
MMLU62.5%#60 of 81, top 75%Epoch AI
MMLU62.5%#60 of 81, top 75%Epoch AI
MMLU59.9%Epoch AI
OpenBookQA79.8%#7 of 19, top 37%Epoch AI
TriviaQA75.2%#15 of 25, top 60%Epoch AI

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English944LMArena2026-10-08
LMArena Non-English1012#283 of 297, top 96%LMArena2026-10-08
LMArena Chinese1009#277 of 285, top 98%LMArena2026-10-08
LMArena Chinese932LMArena2026-10-08
LMArena French946LMArena2026-10-08
LMArena French1037#221 of 223, top 100%LMArena2026-10-08
LMArena German987#228 of 231, top 99%LMArena2026-10-08
LMArena German931LMArena2026-10-08
LMArena Japanese878#211 of 211, top 100%LMArena2026-10-08
LMArena Russian1001LMArena2026-10-08
LMArena Russian1018#273 of 283, top 97%LMArena2026-10-08
LMArena Spanish1026#225 of 226, top 100%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1003LMArena2026-10-08
LMArena Instruction Following1060#276 of 298, top 93%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1060#276 of 291, top 95%LMArena2026-10-08
LMArena Longer Query1001LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1024LMArena2026-10-08
LMArena Text1090#276 of 297, top 93%LMArena2026-10-08
LMArena Creative Writing1028LMArena2026-10-08
LMArena Creative Writing1068#275 of 295, top 94%LMArena2026-10-08
LMArena Multi-Turn1008LMArena2026-10-08
LMArena Multi-Turn1062#273 of 295, top 93%LMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
mistral$0.25$0.25—2026-10-10

All Mistral AI API prices → Estimate your cost →

Compare Mistral 7B

Other Mistral AI models

Frequently asked questions

How good is Mistral 7B?

Mistral 7B by Mistral AI ranks 351st of 354 ranked models on the Noometry Index as of October 2026, with a score of 23.0. Its strongest category is long context, where it ranks 271st. API pricing starts at $0.25 per million input tokens and $0.25 per million output tokens, with a 8K-token context window.

How much does Mistral 7B cost?

Mistral 7B costs $0.25 per million input tokens and $0.25 per million output tokens on Mistral AI's own API.

What is Mistral 7B's context window?

Mistral 7B accepts up to 8K tokens of input and can write up to 8K tokens in one response.

Is Mistral 7B open source?

Yes. Mistral 7B's weights are downloadable; check the license for commercial terms.

What are Mistral 7B's strengths and weaknesses?

Relative to other ranked models, Mistral 7B places best in long context, writing & preference, instruction following and lowest in math, knowledge, reasoning.

What is Mistral 7B best at?

Its best category is long context, where it ranks 271st on Noometry.