token.app › Compare › Codestral 2508 vs Inkling Small
Codestral 2508 vs Inkling Small
Codestral 2508 is 1.4× cheaper per token on a blended 3:1 basis. Inkling Small accepts the larger context window (524K).
Head to head
Blended price uses a 3:1 input:output mix
| Codestral 2508 | Inkling Small | |
|---|---|---|
| Provider | Mistral AI | Thinkingmachines |
| Input $/1M | $0.300 | $0.450 |
| Output $/1M | $0.900 | $1.20 |
| Blended $/1M | $0.450 | $0.637 |
| Context window | 256K | 524K |
| Max output | — | 262K |
| Released | Aug 2025 | Jul 2026 |
| Licence | Proprietary | Open weights |
Choose Codestral 2508 if…
- Costs less per token — $0.450 vs $0.637 blended
Choose Inkling Small if…
- Larger context window (524K)
- Open weights — self-hostable
- Newer model (released Jul 2026)
FAQ
Which is better, Codestral 2508 or Inkling Small?
We do not hold independently-run benchmark scores covering both Codestral 2508 and Inkling Small, so we make no quality claim. Their pricing and specifications are compared on this page.
Is Codestral 2508 cheaper than Inkling Small?
Codestral 2508 is cheaper: $0.450 versus $0.637 per million tokens blended at 3:1 input:output. Input alone: $0.300 vs $0.450. Output alone: $0.900 vs $1.20.
What context windows do Codestral 2508 and Inkling Small have?
Codestral 2508: 256,000 tokens. Inkling Small: 524,288 tokens.
Who makes Codestral 2508 and Inkling Small?
Codestral 2508 is made by Mistral AI. Inkling Small is made by Thinkingmachines.