APIs
Skill Proficiency Analysis #
Extract, score, and band skills across documents. Priced per API call, with payload size determining the charge.
Pricing construct
- Base rate
- $2.50 / 25 credits per API call
- Base payload included
- Up to 1 MB of input payload
- Additional payload
- $2.50 (25 credits) per additional 1 MB
- Maximum payload size
- 10 MB per API call
- Multiple files
- Multiple files may be uploaded in one API call. The combined size determines the charge.
- Compressed files
- For ZIP / compressed formats, the uncompressed size is used for billing.
- Supported formats
.pdf.doc.docx.zip
Pricing tiers
| Payload size | Cost (USD) | Cost (Credits) |
|---|---|---|
| Up to 1 MB | $2.50 | 25 |
| 1 - 2 MB | $5.00 | 50 |
| 2 - 3 MB | $7.50 | 75 |
| 3 - 4 MB | $10.00 | 100 |
| 4 - 5 MB | $12.50 | 125 |
| 5 - 6 MB | $15.00 | 150 |
| 6 - 7 MB | $17.50 | 175 |
| 7 - 8 MB | $20.00 | 200 |
| 8 - 9 MB | $22.50 | 225 |
| 9 - 10 MB (max) | $25.00 | 250 |
Note: Payload size is rounded up to the next whole MB for billing. For ZIP / compressed files, the uncompressed size determines the applicable tier. Failed calls are not charged.
Call Quality Assessment #
Score call-centre transcripts across nine service-excellence dimensions and return a grade, dimension breakdown, sentiment delta, and an AI-coached rewrite. Priced per transcript, by input type and language.
Pricing construct
- Call Transcript Text · English
- Score ($0.50 = 5 credits) per MB of transcript
- Call Transcript Text · Non-English
- Score ($0.70 = 7 credits) per MB of transcript
- Call Audio · English or Non-English
- $0.30 = 3 credits per Minute Duration of Transcript
- Payload size
- Up to 1 MB of input payload
- Supported formats
.txt.json.html- Languages
- English, Hindi, Tamil, Telugu, Kannada, Malayalam, Marathi, Punjabi.
Note: Failed calls are not charged.
Agents
Skill Proficiency Analyzer #
Chat agent that reads documents, identifies every skill, and bands each one Expert / Advanced / Proficient / Beginner. Priced per MB of input.
Pricing construct
- Base rate
- $2.50 / 25 credits per MB of input
- Billing
- Input size is rounded up to the next whole MB. A document up to 1 MB is one unit; each additional MB adds $2.50.
- Conversation rate
- $0.20 / 2 credits per conversation turn in Ask Anything
- Bulk & ZIP
- Billed on the total uncompressed size of all files in the batch. Each document is analysed as its own profile.
- Failed files
- Charged on success only - you pay for successfully processed data, never for files that fail.
- Supported formats
.pdf.doc.docx.zip- Access
- Workspace-scoped. Documents aren't retained after processing.
Price by input size
| Input size | Cost (USD) | Cost (Credits) |
|---|---|---|
| Up to 1 MB | $2.50 | 25 |
| 1 - 2 MB | $5.00 | 50 |
| 2 - 3 MB | $7.50 | 75 |
| 3 - 4 MB | $10.00 | 100 |
| 4 - 5 MB | $12.50 | 125 |
| 5 - 6 MB | $15.00 | 150 |
| 6 - 7 MB | $17.50 | 175 |
| 7 - 8 MB | $20.00 | 200 |
| 8 - 9 MB | $22.50 | 225 |
| 9 - 10 MB | $25.00 | 250 |
Note: Input size is rounded up to the next whole MB. For ZIP / bulk uploads, the uncompressed size determines the tier. Larger inputs continue at $2.50 per additional MB. Failed files are not charged.
Call Quality Assessment #
Scores customer-care call transcripts across nine service-excellence dimensions, returns a grade with full dimension breakdown, and surfaces AI-coached rewrites to improve agent performance. Priced per completed run, by input type and language.
Pricing construct
- Call Transcript Text · English
- Score ($0.50 = 5 credits) per MB of transcript
- Call Transcript Text · Non-English
- Score ($0.70 = 7 credits) per MB of transcript
- Call Audio · English or Non-English
- $0.30 = 3 credits per Minute Duration of Transcript
- Coached Rewrite (Add On)
- $0.70 = 7 credits per MB of transcript
- Payload size
- Up to 1 MB of input payload
- What counts as a run
- One transcript scored end-to-end: nine-dimension analysis, grade (A+ to E), sentiment delta, and AI-coached rewrite.
- Languages
- English, Hindi, Tamil, Telugu, Kannada, Malayalam, Marathi, Punjabi.
- Failed runs
- Runs that fail before producing a result are not charged.
- Access
- Workspace-scoped. Each transcript is processed in isolation.
Note: Failed runs are neither charged nor counted.