Mistral 7B Instruct v0.3
May 2024
Mistral 7B Instruct v0.3Compare inon
Mistral 7B Instruct v0.3 measured on 3 of 104 Chat & Knowledge benchmarks3
BenchmarkByScoreRank
Legacy · saturated or no longer reported2
No benchmark matches
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperLlama 3.1 8B InstructunstatedOpen×6.4-28%$0.02 / $0.04Memory (4-bit): 3.6 GB → 4 GB
BetterGPT-5 (2025-08-07)high×27.3+9,864%$1.25 / $10
OpenPhi 4unstated×18.1+103%$0.056 / $0.11Context: 33K → 16K+1CheaperMeta Llama 3 8B InstructOpen+70%-6%$0.03 / $0.04Context: 33K → 8K+1BetterGPT-5 Mini (2025-08-07)high×27.2+1,893%$0.25 / $2ModelScoreCostPer taskWhat changes
Llama 3.1 8B InstructunstatedOpen×6.4-28%$0.02 / $0.04Memory (4-bit): 3.6 GB → 4 GBMeta Llama 3 8B InstructOpen+70%-6%$0.03 / $0.04Context: 33K → 8K+1ModelScoreCostPer taskWhat changes
GPT-5 (2025-08-07)high×27.3+9,864%$1.25 / $10GPT-5 Mini (2025-08-07)high×27.2+1,893%$0.25 / $2o3 (2025-04-16)high×27.2+10,045%$2 / $8
R1 0528unstatedOpen×26.9+1,914%$0.39 / $1.63Memory (4-bit): 3.6 GB → 411 GBGPT-5 Nano (2025-08-07)medium×26.5+299%$0.05 / $0.40ModelScoreCostPer taskWhat changes
Phi 4unstatedOpen×18.1+103%$0.056 / $0.11Context: 33K → 16K+1Llama 3.1 8B InstructunstatedOpen×6.4-28%$0.02 / $0.04Memory (4-bit): 3.6 GB → 4 GB
Gemma 2 9B ItOpen×5.8+9%$0.03 / $0.06Context: 33K → 8K+1Phi 3 Medium 128K InstructOpen×4.9——Memory (4-bit): 3.6 GB → 7 GBMeta Llama 3 8B InstructOpen+70%-6%$0.03 / $0.04Context: 33K → 8K+1ModelScoreCostPer taskWhat changes
Mixtral 8x7B Instruct v0.1Open×2.6—$0.54 / $0.54Memory (4-bit): 3.6 GB → 28 GBMistral Large 4thinkingn/a—$1.13Infinite whitespace stream+4Mistral Medium 3.5highOpenn/a—$0.50Memory (4-bit): 3.6 GB → 77 GBMistral Small 4thinkingOpenn/a—$0.015Memory (4-bit): 3.6 GB → 72 GBMinistral 3 14B 2512non-thinkingOpenn/a—$0.019Memory (4-bit): 3.6 GB → 7 GBMinistral 3 3B 2512Openn/a—$0.008Ministral 3 8B 2512Openn/a—$0.011Memory (4-bit): 3.6 GB → 4.5 GBMistral Large 3non-thinkingOpenn/a—$0.031Memory (4-bit): not verifiedAll Mistral models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · 8-bit✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
Benchmarks
in Chat & KnowledgeGPQA Diamond · Epoch15.2#131
MATH Level 53.6#34
OTIS Mock AIME 2024 20250.3#128
FAQ
How much does Mistral 7B Instruct v0.3 cost?›
$0.028 per 1M input tokens and $0.054 per 1M output tokens (cheapest host: DeepInfra).
Is Mistral 7B Instruct v0.3 free to use?›
There is no free API; the cheapest paid price is $0.028 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Mistral 7B Instruct v0.3, and what is it best at?›
#131 of 133 in Chat & Knowledge.
Is Mistral 7B Instruct v0.3 better than Meta Llama 3 8B Instruct?›
On MATH Level 5, Mistral 7B Instruct v0.3 scores 3.6 vs 6.1 — lower.
How can I use Mistral 7B Instruct v0.3?›
Through the API of DeepInfra.
What is the context window of Mistral 7B Instruct v0.3?›
33K tokens.
When was Mistral 7B Instruct v0.3 released? Is there a newer version?›
Released 2024-05-22. It is the current version in its line.
Is Mistral 7B Instruct v0.3 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Mistral 7B Instruct v0.3 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Mistral 7B Instruct v0.3 read images?›
No, it accepts text only.
What is Mistral 7B Instruct v0.3?›
Mistral 7B Instruct v0.3 is an open-weights language model from Mistral, released May 2024.
What are the best alternatives to Mistral 7B Instruct v0.3?›
Llama 3.1 8B Instruct (cheaper), GPT-5 (2025-08-07) (better), Phi 4 (open weights).