Codestral vs Gemini 2.5 Flash: Detailed Comparison
Choosing between Codestral (Mistral AI) and
Gemini 2.5 Flash (Google) comes down to three things:
per-token pricing, context window, and which capability matters most for your workload.
Codestral costs $0.20/M input vs
$0.30/M for Gemini 2.5 Flash;
context windows are 32K vs
1.0M tokens. Detailed breakdown below.
Side-by-side specs
| Spec | Codestral | Gemini 2.5 Flash |
|---|
| Provider | Mistral AI | Google |
|---|
| Released | 2024-05-29 | 2025-04-09 |
|---|
| Input price |
$0.20/M |
$0.30/M |
| Output price |
$0.60/M |
$2.50/M |
| Cached input |
— |
$0.0750/M |
| Context window |
32K |
1.0M |
| Max output |
4K |
66K |
| Modalities |
text |
text image audio video |
| Tokenizer |
mistral |
gemini |
Capability matrix
| Capability | Codestral | Gemini 2.5 Flash |
|---|
| function calling |
Yes |
Yes |
| json mode |
Yes |
Yes |
| streaming |
Yes |
Yes |
| code |
Yes |
No |
| vision |
No |
Yes |
| audio |
No |
Yes |
| video |
No |
Yes |
Per-call cost on typical workloads
| Workload (in/out tokens) | Codestral | Gemini 2.5 Flash | Cheaper by |
|---|
| Standard chat (1K / 500) |
$0.000500 |
$0.001550 |
Codestral by $0.001050 |
| RAG (4K / 500) |
$0.001100 |
$0.002450 |
Codestral by $0.001350 |
| Long doc (20K / 1K) |
$0.004600 |
$0.008500 |
Codestral by $0.003900 |
| Very long context (100K / 2K) |
$0.020900 |
$0.033750 |
Codestral by $0.012850 |
When to choose Codestral over Gemini 2.5 Flash
- Per-token input cost is 33% lower — meaningful for high-volume workloads.
- Supports code — Gemini 2.5 Flash does not.
When to choose Gemini 2.5 Flash over Codestral
- Larger context window (1.0M vs 32K).
- Supports vision — Codestral does not.
- Supports audio — Codestral does not.
- Supports video — Codestral does not.
Related comparisons