MDL
Mistral 7B Instruct
Mistral
Verified against Artificial Analysis · Jul 13, 2026
Capability
14
percentile index
Price /M
$0.25 / $0.25
in / out
Context
—
Avg score
9.5
9 benchmarks
AA Intelligence
2.1
index
Speed
105 tok/s
TTFT 0.4s
Released
Sep 27, 2023
About
Mistral 7B Instruct is a open-weights model in the Mistral family from Mistral. Benchmarked on 9 evals, averaging 9.5.
Benchmark scores · 9
avg 9.5 — higher is betterThe research behind this model
all papers mentioning it →Bag of Dims: Training-Free Mechanistic Interpretability via Dimension-Level Sign Patterns▲ 2 on HF · Jun 17, 2026Mistral 7B282 cites · Oct 10, 2023WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instruct27 cites · Aug 18, 2023Compact Language Models via Pruning and Knowledge Distillation10 cites · Jul 19, 2024Falcon Mamba: The First Competitive Attention-free 7B Language Model3 cites · Oct 7, 2024
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed105 tok/s
Latency (TTFT)0.4s
Price in / out$0.25 / $0.25 per 1M
Blended price$0.25 / 1M
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$0.25/M