Baidu lists 5 models and Scaledown 2. Scaledown is cheaper on a typical model ($0.03 per 1M tokens median). Below: line-up, price tiers, the cheapest model for each capability, and what each costs for five real workloads.
| Baidu | Scaledown | |
|---|---|---|
| Models listed | 5 | 2 |
| Models with published pricing | 5 | 2 |
| Typical price (median, 3:1 blend per 1M) | $0.25 | $0.03 |
| Cheapest paid model | Baidu: ERNIE 4.5 21B A3B ($0.12) | ScaleDown: Extract ($0.03) |
| Longest context | Baidu: ERNIE 4.5 21B A3B Thinking (131K) | ScaleDown: Extract (1M) |
| Free models | 0 | 0 |
| Newest model | Baidu: ERNIE 4.5 21B A3B Thinking (2025-10-09) | ScaleDown: Extract (2026-10-13) |
| Price changes in the last 90 days | 0 | 0 |
For each workload: the cheapest model from each provider that can actually do it (fits the context window; tool calling required for the agent), and its monthly cost at list price.
| Workload (per month) | Baidu | Scaledown |
|---|---|---|
| Customer-support chatbot 50,000 conversations — each a 6-turn conversation | $10.85 with Baidu: ERNIE 4.5 21B A3B | $3.00 with ScaleDown: Extract |
| RAG search over documents 20,000 questions — each a question answered from ~8 retrieved passages | $11.20 with Baidu: ERNIE 4.5 21B A3B | $4.80 with ScaleDown: Extract |
| Coding / tool-using agent 2,000 tasks — each a multi-step task with tool calls and file context | $11.76 with Baidu: ERNIE 4.5 21B A3B | no suitable model |
| Batch summarisation 10,000 documents — each a 6,000-word report summarised to a page | $7.28 with Baidu: ERNIE 4.5 21B A3B | $3.20 with ScaleDown: Extract |
| Long-document analysis 500 documents — each a 150-page contract or codebase read in one pass | $3.78 with Baidu: ERNIE 4.5 21B A3B | $2.00 with ScaleDown: Extract |
| You need | Baidu | Scaledown |
|---|---|---|
| Any model | Baidu: ERNIE 4.5 21B A3B ($0.12) | ScaleDown: Extract ($0.03) |
| With tool calling | Baidu: ERNIE 4.5 21B A3B ($0.12) | none listed |
| With image input | Baidu: ERNIE 4.5 VL 28B A3B ($0.25) | ScaleDown: Extract ($0.03) |
| With reasoning mode | Baidu: ERNIE 4.5 21B A3B Thinking ($0.12) | none listed |
| 128K+ context | Baidu: ERNIE 4.5 21B A3B Thinking ($0.12) | ScaleDown: Extract ($0.03) |
| 1M+ context | none listed | ScaleDown: Extract ($0.03) |
| Model type | Baidu | Scaledown |
|---|---|---|
| text+image+file->text | 0 | 2 |
| text+image->text | 2 | 0 |
| text->text | 3 | 0 |
| Price per 1M (3:1 blend) | Baidu | Scaledown |
|---|---|---|
| Free | 0 | 0 |
| Under $0.50 | 4 | 2 |
| $0.50 – $3 | 1 | 0 |
| $3 – $10 | 0 | 0 |
| $10 and up | 0 | 0 |
Scaledown is cheaper on a typical model: a median of $0.03 per 1M tokens (3:1 input/output) against $0.25 for Baidu. The cheapest paid models are Baidu: ERNIE 4.5 21B A3B at $0.12 and ScaleDown: Extract at $0.03.
It depends on the job: Baidu offers the broader line-up (5 models vs 2), Scaledown the lower typical price, Scaledown the longest context (1M tokens). The workload table above shows which wins for chat, RAG, agents, batch jobs and long documents.
Baidu: Baidu lists 5 models and Scaledown lists 2.
Scaledown: Baidu's longest is Baidu: ERNIE 4.5 21B A3B Thinking at 131K tokens; Scaledown's is ScaleDown: Extract at 1M.
Neither lists a free model; both are pay-per-token.
Baidu: 2 of 5 models accept images. Scaledown: 2 of 2.