Mistral Small 4 vs Nova Lite
Mistral vs Amazon · Prices checked 12 August 2026
On price Nova Lite edges it, $0.114 per million tokens blended versus $0.160, about 29 per cent less. A single 100K-input, 10K-output job costs $0.0084 on Nova Lite against $0.013 on Mistral Small 4.
Capability favours Nova Lite, ahead on 2 of 3 suites they both publish. The widest gap is on HumanEval, where Nova Lite leads by 12 points. Nova Lite reads more in one request: 300,000 tokens of context against 128,000.
Pricing and capacity
| Metric | Mistral Small 4 | Nova Lite | Difference |
|---|---|---|---|
| Input / 1M tokens | $0.100 | $0.060 | 40% |
| Output / 1M tokens | $0.300 | $0.240 | 20% |
| Blended · 70/30calculated | $0.160 | $0.114 | 29% |
| Blended · 50/50calculated | $0.200 | $0.150 | 25% |
| Blended · 80/20calculated | $0.140 | $0.096 | 31% |
| Blended · 20/80calculated | $0.260 | $0.204 | 22% |
| Job cost · 100K + 10Kcalculated | $0.013 | $0.0084 | 35% |
| Job cost · 1M + 100Kcalculated | $0.130 | $0.084 | 35% |
| Context window | 128,000 | 300,000 | 2.3× |
Input and output rates are as published by each provider. Blended and job-cost rows are calculated comparison metrics, not published prices — how they are derived.
Benchmarks
| Benchmark | Mistral Small 4 | Nova Lite | Leader |
|---|---|---|---|
| MMLU | 71% | 75% | Nova Lite |
| HumanEval | 68% | 80% | Nova Lite |
| LMArena Elo | 1175 Elo | 1170 Elo | Mistral Small 4 |
Only suites both models report are listed, so a missing score is never shown as a loss. Figures are provider-published or from public leaderboards and are not independently re-run. Where a model has no score of its own it inherits its base model’s, since the weights are the same.
What each model is for
Mistral Small 4
Mistral · Efficient
Mistral's 24B small model: open-weights, fast, and strong on instruction following for its size.
- Open weights (Apache 2.0)
- Single-GPU friendly
- 128K context
Frequently asked questions
Which is cheaper, Mistral Small 4 or Nova Lite?
On price Nova Lite edges it, $0.114 per million tokens blended versus $0.160, about 29 per cent less. A single 100K-input, 10K-output job costs $0.0084 on Nova Lite against $0.013 on Mistral Small 4. Blended cost weights input at 70% and output at 30%; price your own token split for an exact answer, since the cheaper model on a blended basis is not always cheaper for your workload.
Which performs better, Mistral Small 4 or Nova Lite?
Capability favours Nova Lite, ahead on 2 of 3 suites they both publish. The widest gap is on HumanEval, where Nova Lite leads by 12 points. These are the figures each provider or public leaderboard publishes, not results we re-run, and a benchmark average can hide the specific behaviour your workload depends on.
Which has the larger context window, Mistral Small 4 or Nova Lite?
Nova Lite reads more in one request: 300,000 tokens of context against 128,000. The context window caps everything in a single request: prompt, retrieved documents, conversation history and the model’s own output all share it.
Can I compare Mistral Small 4 and Nova Lite on my own numbers?
Yes. Open both in the side-by-side comparison tool to see every field at once, or use the token calculator to price your actual prompt and monthly request volume against each model. Both are free and need no sign-up.
Price it for your own workload
Related matchups
Terms used on this page
Prices and scores are recomputed on every deploy from the catalogue. Prices checked 12 August 2026.