Models, measured side by side
DeepSeek V3 vs Mistral Large
Compare DeepSeek V3, Mistral Large by token pricing, context length, benchmark results, speed and tool support.
Monthly workload
Monthly cost breakdown
InputOutput
DeepSeek V3
$10 / mo
Mistral Large
$70 / mo
Performance comparison
DeepSeek V3
8.5
Mistral Large
5.8
Overview
| Overview | DeepSeek V3 | Mistral Large |
|---|---|---|
| Provider | DeepSeek | Mistral AI |
| Released | 2024-12 | 2024-11 |
| Status | Active | Active |
| Input formats | text | textfile |
| Output formats | text | text |
Provider
- DeepSeek V3
- DeepSeek
- Mistral Large
- Mistral AI
Released
- DeepSeek V3
- 2024-12
- Mistral Large
- 2024-11
Status
- DeepSeek V3
- Active
- Mistral Large
- Active
Input formats
- DeepSeek V3
- text
- Mistral Large
- textfile
Output formats
- DeepSeek V3
- text
- Mistral Large
- text
Pricing & capacity
| Pricing & capacity | DeepSeek V3 | Mistral Large |
|---|---|---|
| Input / 1M tokens | $0.26 | $2.00 |
| Output / 1M tokens | $1.03 | $6.00 |
| Cached input / 1M | Not listed | $0.20 |
| Monthly cost | $10 | $70 |
| Context window | 164K | 128K |
| Maximum output | 16K | 102K |
Input / 1M tokens
- DeepSeek V3
- $0.26
- Mistral Large
- $2.00
Output / 1M tokens
- DeepSeek V3
- $1.03
- Mistral Large
- $6.00
Cached input / 1M
- DeepSeek V3
- Not listed
- Mistral Large
- $0.20
Monthly cost
- DeepSeek V3
- $10
- Mistral Large
- $70
Context window
- DeepSeek V3
- 164K
- Mistral Large
- 128K
Maximum output
- DeepSeek V3
- 16K
- Mistral Large
- 102K
Benchmarks & performance
| Benchmarks & performance | DeepSeek V3 | Mistral Large |
|---|---|---|
| Arena Elo | 1,211 | 1,189 |
| MMLU-Pro | 75.4% | 71.8% |
| Intelligence Index | 8.5 | 5.8 |
| GPQA | 55.7% | 35.1% |
| Humanity’s Last Exam | 2.9% | 3.5% |
| SciCode | 35.8% | Not listed |
Arena Elo
- DeepSeek V3
- 1,211
- Mistral Large
- 1,189
MMLU-Pro
- DeepSeek V3
- 75.4%
- Mistral Large
- 71.8%
Intelligence Index
- DeepSeek V3
- 8.5
- Mistral Large
- 5.8
GPQA
- DeepSeek V3
- 55.7%
- Mistral Large
- 35.1%
Humanity’s Last Exam
- DeepSeek V3
- 2.9%
- Mistral Large
- 3.5%
SciCode
- DeepSeek V3
- 35.8%
- Mistral Large
- Not listed
Tools & features
| Tools & features | DeepSeek V3 | Mistral Large |
|---|---|---|
| Function calling | Supported | Supported |
| Structured outputs | Supported | Supported |
| JSON mode | Supported | Supported |
| Deterministic seed | Supported | Supported |
| Prompt caching | Not listed | Supported |
Function calling
- DeepSeek V3
- Supported
- Mistral Large
- Supported
Structured outputs
- DeepSeek V3
- Supported
- Mistral Large
- Supported
JSON mode
- DeepSeek V3
- Supported
- Mistral Large
- Supported
Deterministic seed
- DeepSeek V3
- Supported
- Mistral Large
- Supported
Prompt caching
- DeepSeek V3
- Not listed
- Mistral Large
- Supported
API & availability
| API & availability | DeepSeek V3 | Mistral Large |
|---|---|---|
| API identifier | deepseek-v3 | mistral-large |
| API providers | StreamLake, DeepInfra | Mistral, Mistral |
| Knowledge cutoff | 2024-07-31 | 2024-11-30 |
| Prices checked | Oct 8, 2026 | Oct 8, 2026 |
API identifier
- DeepSeek V3
deepseek-v3- Mistral Large
mistral-large
API providers
- DeepSeek V3
- StreamLake, DeepInfra
- Mistral Large
- Mistral, Mistral
Knowledge cutoff
- DeepSeek V3
- 2024-07-31
- Mistral Large
- 2024-11-30
Prices checked
- DeepSeek V3
- Oct 8, 2026
- Mistral Large
- Oct 8, 2026
About the models
Comparison FAQ
DeepSeek V3: $0.26 input and $1.03 output per million tokens. Mistral Large: $2.00 input and $6.00 output per million tokens. The cheaper choice depends on your input-to-output ratio; a model can have a lower input rate but a higher output rate.