AI Rate CardDeepSeek V4.1 Flash vs Llama-3.2-90B-Vision-Instruct: API price and specs
DeepSeek V4.1 Flash is 87% cheaper than Llama-3.2-90B-Vision-Instruct on input price.
DeepSeek V4.1 Flash is 62% cheaper than Llama-3.2-90B-Vision-Instruct on output price.
DeepSeek V4.1 Flash is 83% cheaper than Llama-3.2-90B-Vision-Instruct on a 1M-in / 200k-out workload.
DeepSeek V4.1 Flash has a 700% larger context than Llama-3.2-90B-Vision-Instruct.
| Field | DeepSeek V4.1 Flash | Llama-3.2-90B-Vision-Instruct |
|---|---|---|
| Lowest input, per 1M | $0.045 | $0.35 +677.8% |
| Median input, per 1M | $0.30 | $0.35 +16.7% |
| Lowest output, per 1M | $0.15 | $0.40 +166.7% |
| Median output, per 1M | $1.20 +200% | $0.40 |
| Lowest unit price | — | — |
| Cost of 1M in + 200k out | $0.075 | $0.43 +473.3% |
| Context | 1M | 131K +87.5% |
| Max output | 1M | 8K +99.2% |
| Providers | 60 | 6 +90% |
| Regions | APAC, EU, GLOBAL, US | — |
| Accepts | application/pdf, file, image, image/gif, image/jpeg, image/png, image/webp, pdf, text | image, text |
| Returns | text | text |
| Type | Text models | Text models |
| Free option | yes | yes |
| Cheapest provider | openrouter | deepinfra |
Your own workload: how many tokens the model gets (input) and gives back (output) - for example, in a month. The "Cost" row of the table above is recalculated for these numbers, for each model.
Computed from current offers each time the page opens; the best value of a row is bold, the percent is how far behind the best.
Providers selling both
aihubmix, deepinfra, io-net, nvidia, requesty, trustedrouter
Frequently asked questions
Which is cheaper, DeepSeek V4.1 Flash or Llama-3.2-90B-Vision-Instruct?
DeepSeek V4.1 Flash is 87% cheaper than Llama-3.2-90B-Vision-Instruct on input price. DeepSeek V4.1 Flash is 62% cheaper than Llama-3.2-90B-Vision-Instruct on output price. DeepSeek V4.1 Flash is 83% cheaper than Llama-3.2-90B-Vision-Instruct on a 1M-in / 200k-out workload.
Which has the larger context, DeepSeek V4.1 Flash or Llama-3.2-90B-Vision-Instruct?
DeepSeek V4.1 Flash: 1M tokens; Llama-3.2-90B-Vision-Instruct: 131K tokens.
Where can I use both DeepSeek V4.1 Flash and Llama-3.2-90B-Vision-Instruct?
aihubmix, deepinfra, io-net, nvidia, requesty, trustedrouter