# TensorFeed AI Research Milestones Feed

> TensorFeed AI Research Milestones Feed is a paid API for AI agents from tensorfeed.ai, paid per call via x402, $0.02/call, status unknown (last checked 2026-09-15).

Returns arXiv preprints from the last 30 days flagged as milestone candidates, with structured reasoning identifying benchmark improvements, model releases, or novel architectures extracted by an offline Qwen 3.6 27B model.

## Facts

- Endpoint: GET https://tensorfeed.ai/api/premium/research/milestones
- Price: $0.02/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/tensorfeed-ai-51c903e9
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_IwCIRRVz8eg-mpIO9hkHq

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability tensorfeed-ai-51c903e9
```

Example prompt: Pull the latest AI research milestones from the last 30 days — I want the arXiv papers flagged as genuine breakthroughs, with the benchmark name, how much it improved, and the reasoning for why each one qualifies.

## When to prefer this

Use this endpoint when you need a curated, conservatively filtered list of genuinely significant AI research milestones rather than raw paper feeds or citation-ranked lists. Prefer this over general arXiv search when you want pre-extracted benchmark deltas and structured milestone reasoning, not just keyword matches. Ideal for automated research digests, competitive intelligence on model releases, or staying current on architectural innovations without manual review.

## Known failure modes

- Payment not included or insufficient USDC balance returns 402 Payment Required
- No milestones found in the 30-day window returns empty list
- Upstream arXiv data lag may mean very recent papers are not yet processed
- Qwen extraction model offline or stale may reduce coverage
- Rate limiting may apply at high call frequency

## How this service works

Last 30 days of arXiv preprints flagged as milestone candidates by an offline Qwen 3.6 27B per-paper extraction pass. Each paper carries structured reasoning naming the benchmark + quantified delta, model release, or novel architecture justification. Conservative; false positives are worse than false negatives.

## Output

A list of arXiv preprints from the last 30 days flagged as milestone candidates, each with structured reasoning that names the specific benchmark and quantified delta, identifies model releases, or justifies novel architecture significance — conservatively filtered so false positives are minimized.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "method"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "http"
    },
    "method": {
     "enum": [
      "GET"
     ],
     "type": "string"
    },
    "queryParams": {
     "type": "object"
    }
   },
   "additionalProperties": false
  },
  "output": {
   "type": "object",
   "required": [
    "type"
   ],
   "properties": {
    "type": {
     "type": "string"
    },
    "example": {
     "type": "object"
    }
   }
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/tensorfeed-ai-51c903e9/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from tensorfeed.ai](https://www.zero.xyz/host/tensorfeed.ai/llms.txt)