Compare · ModelsLive · 2 picked · head to head
GPT-4 (older v0314) vs Mixtral 8x7B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-4 (older v0314) wins on 4/4 benchmarks
GPT-4 (older v0314) wins 4 of 4 shared benchmarks. Leads in knowledge · math.
Category leads
knowledge·GPT-4 (older v0314)math·GPT-4 (older v0314)
Hype vs Reality
Attention vs performance
GPT-4 (older v0314)
#81 by perf·no signal
Mixtral 8x7B
#85 by perf·no signal
Best value
GPT-4 (older v0314)
GPT-4 (older v0314)
1.2 pts/$
$45.00/M
Mixtral 8x7B
n/a
no price
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Mistral AI
$14.0B·Tier 1
Head to head
4 benchmarks · 2 models
GPT-4 (older v0314)Mixtral 8x7B
GPQA diamond
GPT-4 (older v0314) leads by +6.9
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-4 (older v0314)
14.3
Mixtral 8x7B
7.5
GSM8K
GPT-4 (older v0314) leads by +17.6
Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve.
GPT-4 (older v0314)
92.0
Mixtral 8x7B
74.4
MMLU
GPT-4 (older v0314) leads by +21.1
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
GPT-4 (older v0314)
81.9
Mixtral 8x7B
60.8
Winogrande
GPT-4 (older v0314) leads by +20.6
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
GPT-4 (older v0314)
75.0
Mixtral 8x7B
54.4
Full benchmark table
| Benchmark | GPT-4 (older v0314) | Mixtral 8x7B |
|---|---|---|
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 14.3 | 7.5 |
GSM8K Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve. | 92.0 | 74.4 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 81.9 | 60.8 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 75.0 | 54.4 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $30.00 | $60.00 | 8K tokens (~4 books) | $375.00 | |
| — | — | — | — |