Papers
Overview
arXiv full text, structured section by section with a TLDR for each; bioRxiv and medRxiv alongside it.
- status
- live
- document id
arxiv_ide.g.2409.05591- citation
[arXiv:2409.05591]- updated
- 2026-09
Common calls
Find papers
POST /v1/papers/search {"query": "speculative decoding", "filters": {"publish_at": {"gte": "2024"}, "categories": ["cs.CL"]}}Read a paper
GET /v1/papers/read/2409.05591Just ask
POST /v1/papers/ask {"query": "DEER 在 HumanEval 上加速多少", "effort": "low"}The query matters more than the parameters. Be specific — “what compression ratio does KV cache eviction report on LongBench” beats “kv cache”; ask for numbers if you want numbers. Chinese works directly. Year, venue, category, author, institution and minimum citations belong in filters. If the results miss, rephrase — more effort only adds reading rounds.
The fine print
- `section` (with level=full) must match sections[].name from level=detail exactly, numbering included (e.g. '1. Introduction').
Search
POST
/v1/papers/searchHybrid retrieval, one card per hit. 2 credits. Free sample:
transformer.Filter fields
| field | type · ops | meaning |
|---|---|---|
| authors | keyword · eq any | Author name; any match, also boosts ranking |
| orgs | keyword · eq any | Organisation; any match, also boosts ranking |
| categories | keyword · eq anycs.AI · cs.AR · cs.CC · cs.CE · cs.CG · cs.CL … 117 values, /facets lists them all | arXiv category, e.g. cs.CL |
| publish_at | datetime · range | Publication date range: YYYY / YYYY-MM / YYYY-MM-DD |
| citations | integer · range | Minimum citation count (gte only) |
3 more, all usable
| field | type · ops | meaning |
|---|---|---|
| source | keyword · eqarxiv · biorxiv · medrxiv | arxiv (default) / biorxiv / medrxiv |
| venues | keyword · eq anyNeurIPS · ICML · ICLR · AAAI · IJCAI · AISTATS … 58 values, /facets lists them all | Venue alias, e.g. NeurIPS, ICLR, CVPR (NIPS ↔ NeurIPS resolve) |
| venue_year | integer · eq | Venue year |
Parameters
| Parameter | Type | Description |
|---|---|---|
| query REQUIRED | string | 1–1000 characters. Or an array of up to 5 (batch; hits become results[]). |
| top_k | integer= 10 | 1–100. |
| fields | brief · head= brief | head costs one more credit and returns the head view per hit. |
| filters | json | Filter DSL object (see Getting started); declared fields are listed above. |
| mode | hybrid= hybrid | This domain has one mode — keyword matching, not semantics. |
| offset | integer= 0 | Pagination. |
Example
{ "domain": "papers", "query": "…", "hits": [ { "id": "2409.05591", "score": 0.83, "brief": { … } } ], "total": null, "coverage": …, "warnings": [],
"meta": { "request_id": "req_…", "credits": { "charged": 2, "balance": 1998, "pool": "allowance" }, "latency_ms": 41, "cached": false } } Undeclared filter fields return 400 rather than being ignored silently. Filters combine with AND;
_or / _not are available.Read
GET
/v1/papers/read/{arxiv_id}Three depths.
detail is the default. Free sample: 2409.05591.Levels
| level | credits | returns |
|---|---|---|
| brief | 1 | Title, TLDR, keywords, date, citations, GitHub. ~300 tok |
| detail (default) | 2 | Metadata + section map (name / tldr / tokens per section) + total tokens |
| full | 5 | The whole paper as markdown |
Extras
One side of the same document, in place of a level.
| extra | credits | returns |
|---|---|---|
| trending | 1 | X/Twitter attention signal |
Parameters
| Parameter | Type | Description |
|---|---|---|
| arxiv_id (path) REQUIRED | string | Document id (arxiv_id). |
| level | brief · detail · full= detail | How deep to read. |
| extra | trending | Read one side instead of a level. |
| format | json · md · html= json | Response shape. |
| section | string | (level = full: read one section) |
| characters | integer= 10000 | (level = full: first N characters) |
Example
{ "domain": "papers", "id": "2409.05591", "level": "detail", "view": "head", "data": { … } | "text": "…", "token_count": 1834, "meta": { … } } The older
GET /v1/papers/doc/{id}?view=… still works.Read many
POST
/v1/papers/readUp to 50 ids, one level. Missing ids come back with
error: "not_found" and are not charged.Parameters
| Parameter | Type | Description |
|---|---|---|
| ids REQUIRED | string[] | Document ids. |
| level | brief · detail · full= detail | Level for every id. |
| extra | trending | Read one side instead. |
Example
{ "domain": "papers", "level": "detail", "docs": [ { "id": "…", "data": { … } }, { "id": "…", "error": "not_found" } ], "meta": { … } } Facets
GET
/v1/papers/facets?field={field}Legal values for a filter field, so an agent never guesses; precomputed fields carry counts, declared enums come without. Free, no key.
Parameters
| Parameter | Type | Description |
|---|---|---|
| field REQUIRED | source · categories · venues | Filter field with facets. |
| limit | integer= 50 | Max values. |
Example
{ "domain": "papers", "field": "…", "values": [ { "value": "…", "count": 3120 } ], "meta": { … } } Askregistered key+ /stream
POST
/v1/papers/askHybrid retrieval, reads sections, answers with real [arXiv:id] citations; /stream is NDJSON.
effort low / medium / high = 50 / 100 / 200 credits. Append /stream for NDJSON; the first event is always billing. A failure before the first answer_delta is refunded.Parameters
| Parameter | Type | Description |
|---|---|---|
| query REQUIRED | string | 1–2000 characters. Be specific; ask for numbers if you want numbers. |
| effort | low · medium · high= low | Reading rounds, and the price: 50 / 100 / 200 credits. Your plan caps it. |
| verbose | bool= false | Include the tool-call trace (tool_call / tool_result / thinking events when streaming). |
| top_k | integer= 10 | Speculative prefetch size. |
| stream_answer | bool= true | Streaming only: false emits one answer event instead of deltas. |
| max_rounds | integer | 0–8, overrides the effort preset. |
| force_answer_after | number | Seconds; hard cutover into the answer phase. |
| max_answer_tokens | integer= 4096 | Hitting it sets answer_truncated. |
| language | string | Answer language; defaults to the query’s. |
Example
{ "answer": "… [arXiv:2409.05591] …", "sources": [ … ], "stats": { "rounds": 1, "elapsed_s": 6.2, "answer_truncated": false }, "coverage"?: { … }, "meta": { "credits": { "charged": 50, … } } } Citations are real — every id in
answer exists in this domain (format [arXiv:2409.05591]). sources is the retrieval set, not the citation list. answer_truncated means incomplete.