Compare · ModelsLive · 2 picked · head to head
Mixtral 8x7B vs Mixtral 8x7B Instruct
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Mixtral 8x7B wins on 11/11 benchmarks
Mixtral 8x7B wins 11 of 11 shared benchmarks. Leads in knowledge · math.
Category leads
knowledge·Mixtral 8x7Bmath·Mixtral 8x7B
Hype vs Reality
Attention vs performance
Mixtral 8x7B
#70 by perf·no signal
Mixtral 8x7B Instruct
#71 by perf·no signal
Best value
Mixtral 8x7B Instruct
Mixtral 8x7B
—
no price
Mixtral 8x7B Instruct
107.0 pts/$
$0.54/M
Vendor risk
Who is behind the model
Mistral AI
$14.0B·Tier 1
Mistral AI
$14.0B·Tier 1
Head to head
11 benchmarks · 2 models
Mixtral 8x7BMixtral 8x7B Instruct
ANLI
ANLI (Adversarial NLI) · adversarially constructed natural language inference dataset where each round targets weaknesses found in previous model generations.
Mixtral 8x7B
32.8
Mixtral 8x7B Instruct
32.8
ARC AI2
AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval.
Mixtral 8x7B
83.1
Mixtral 8x7B Instruct
83.1
GPQA diamond
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Mixtral 8x7B
7.5
Mixtral 8x7B Instruct
7.5
GSM8K
Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve.
Mixtral 8x7B
74.4
Mixtral 8x7B Instruct
74.4
HellaSwag
HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios.
Mixtral 8x7B
82.3
Mixtral 8x7B Instruct
82.3
MATH level 5
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Mixtral 8x7B
9.9
Mixtral 8x7B Instruct
9.9
MMLU
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Mixtral 8x7B
60.8
Mixtral 8x7B Instruct
60.8
OpenBookQA
OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting.
Mixtral 8x7B
81.1
Mixtral 8x7B Instruct
81.1
PIQA
PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks.
Mixtral 8x7B
67.2
Mixtral 8x7B Instruct
67.2
TriviaQA
TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents.
Mixtral 8x7B
82.2
Mixtral 8x7B Instruct
82.2
Winogrande
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
Mixtral 8x7B
54.4
Mixtral 8x7B Instruct
54.4
Full benchmark table
| Benchmark | Mixtral 8x7B | Mixtral 8x7B Instruct |
|---|---|---|
ANLI ANLI (Adversarial NLI) · adversarially constructed natural language inference dataset where each round targets weaknesses found in previous model generations. | 32.8 | 32.8 |
ARC AI2 AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval. | 83.1 | 83.1 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 7.5 | 7.5 |
GSM8K Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve. | 74.4 | 74.4 |
HellaSwag HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios. | 82.3 | 82.3 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 9.9 | 9.9 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 60.8 | 60.8 |
OpenBookQA OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting. | 81.1 | 81.1 |
PIQA PIQA (Physical Interaction QA) · tests intuitive physical reasoning by asking models to select the correct approach for everyday physical tasks. | 67.2 | 67.2 |
TriviaQA TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents. | 82.2 | 82.2 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 54.4 | 54.4 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| $0.54 | $0.54 | 33K tokens (~16 books) | $5.40 |