Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | 7B | 6B |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 8k | 32k |
| Benchmarks | ||
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | 26.2% | 9.8% |
HellaSwagReasoning — commonsense completion of short scenarios | 81% | 57% |
MMLUGeneral knowledge — A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. | 62.5% | 45.2% |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | 36.5% | 25.9% |
MATHReasoning — accuracy on the original competition mathematics test set | — | 6.9% |
ARC-ChallengeReasoning — multiple-choice science questions from the AI2 Reasoning Challenge; unrelated to ARC-AGI | 54.9% | — |
WinoGrandeReasoning — resolving ambiguous pronouns in context | 74.2% | — |
| Overview | ||
| Company | Mistral | Z.ai |
| Release date | Sep 27 2023 | Jun 25 2023 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Mistral 7BvsClaude Fable 5.2ChatGLM2-6BvsClaude Fable 5.2Mistral 7BvsGPT-6.2 AstraChatGLM2-6BvsGPT-6.2 AstraMistral 7BvsGemini 4 HydrogenChatGLM2-6BvsGemini 4 HydrogenMistral 7BvsMuse EmberChatGLM2-6BvsMuse EmberMistral 7BvsGrok 4.7ChatGLM2-6BvsGrok 4.7Mistral 7BvsDeepSeek-V4.1-Flash-1017ChatGLM2-6BvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Mistral 7B leads ChatGLM2-6B on 4 of the 4 benchmarks they both report (HumanEval, HellaSwag, MMLU, GSM8K). ChatGLM2-6B shipped 94 days before Mistral 7B, so benchmark comparisons should account for the intervening progress.
Mistral 7B has 7B parameters, while ChatGLM2-6B has 6B. Context windows are 8k (Mistral 7B) vs 32k (ChatGLM2-6B).
On HumanEval, Mistral 7B leads at 26.2% vs ChatGLM2-6B at 9.8%. On HellaSwag, Mistral 7B leads at 81% vs ChatGLM2-6B at 57%. On MMLU, Mistral 7B leads at 62.5% vs ChatGLM2-6B at 45.2%. On GSM8K, Mistral 7B leads at 36.5% vs ChatGLM2-6B at 25.9%.