# ask
[rishi](https://github.com/vedicreader/rishi) picks its backend from the
*shape* of a model id — `litert-community/…` or `.litertlm` goes to
LiteRT, `.gguf` to llama.cpp, `mlx-community/…` to MLX, a hosted name
like `claude-sonnet-5` to fastllm — and raises rather than guessing when
the id says nothing. `MODELS` exists only so the common cases are one
short word.
------------------------------------------------------------------------
source
### resolve_model
``` python
def resolve_model(
model:str=None, runtime:str=None
)->tuple:
```
*`(model_id, runtime)` from a MODELS alias, a full id, or a
`runtime/model` string.*
------------------------------------------------------------------------
source
### cited
``` python
def cited(
answer:str, results
)->L:
```
*The sections an answer actually cited, in citation order — the audit
trail for a claim.*
------------------------------------------------------------------------
source
### split_reasoning
``` python
def split_reasoning(
text:str
)->tuple:
```
*`(answer, thinking)` from a reasoning model’s reply, whichever half of
the tag pair arrived.*
rishi’s `split_think` covers `…` and an opener cut off at
the token cap. MLX chat templates prefill the opener, so a Qwen reply
arrives with a *closing* `` and nothing before it — which
`split_think` leaves alone and
[`cited`](https://vedicreader.github.io/vishalakshi/ask.html#cited)
would then read for citations the answer never makes.
------------------------------------------------------------------------
source
### mk_prompt
``` python
def mk_prompt(
question:str, # what you want to know
ctx, # AttrDict from Vault.context()
max_chars:int=4000, # chars kept per section
related:bool=True, # include the associative leg
)->str:
```
*The user turn: the numbered sections, then the question.*
The numbering is the contract with the model: `[n]` in the answer maps
to `ctx.results[n-1]`, which is what makes an answer checkable against
the vault instead of merely plausible.
------------------------------------------------------------------------
source
### Vault.explain
``` python
def explain(
node_id:str, model:str=None, max_chars:int=6000, **kw
)->AttrDict:
```
*Have a model explain one section in the context of what the vault
connects it to.*
------------------------------------------------------------------------
source
### Vault.ask
``` python
def ask(
question:str, # what you want to know
model:str=None, # a MODELS alias, a full id, or None -> $VISHALAKSHI_MODEL
runtime:str=None, # 'litert' | 'mlx' | 'llama' | 'remote'; inferred from the id if None
sections:int=6, # operative sections retrieved
related:int=6, # associative sections offered as leads
kind:str=None, # restrict retrieval to one or more KINDS
max_chars:int=4000, # chars of each section shown to the model
sp:str="You answer questions from a personal research vault.\n\nYou are given numbered sections retrieved from the user's own corpus — papers, web pages,\ntranscripts, files and their own notes. Answer only from those sections.\n\nRules:\n- Cite every claim with the bracketed number of the section it came from, like [2]. A sentence\n drawing on two sections cites both.\n- If the sections do not answer the question, say exactly what is missing rather than filling the\n gap from memory. A vault that admits a hole is useful; one that guesses is not.\n- Sections marked RELATED were reached by association, not by matching the question. Use them for\n context or to point somewhere worth reading next, and say so when you do.\n- Prefer the user's own notes when they conflict with a source, and flag the disagreement.", # system prompt
fresh:bool=True, # start a new conversation rather than continuing the last
**kw
)->AttrDict: # forwarded to Vault.context
```
*Retrieve, then answer with citations back into the vault.*
[`cited`](https://vedicreader.github.io/vishalakshi/ask.html#cited)
resolves the `[n]` markers in the answer back to `node_id`s you can
`read()`, so every claim is one call away from the text it came from;
`context` is the full retrieval, kept so you can inspect what the model
was and was not shown. A reasoning model’s `` block is split off
into `thinking` rather than left in `answer`: it names sections it then
discards, so leaving it in would put citations in the answer that the
answer does not actually make.
------------------------------------------------------------------------
source
### Vault.chat
``` python
def chat(
model:str=None, runtime:str=None,
sp:str="You answer questions from a personal research vault.\n\nYou are given numbered sections retrieved from the user's own corpus — papers, web pages,\ntranscripts, files and their own notes. Answer only from those sections.\n\nRules:\n- Cite every claim with the bracketed number of the section it came from, like [2]. A sentence\n drawing on two sections cites both.\n- If the sections do not answer the question, say exactly what is missing rather than filling the\n gap from memory. A vault that admits a hole is useful; one that guesses is not.\n- Sections marked RELATED were reached by association, not by matching the question. Use them for\n context or to point somewhere worth reading next, and say so when you do.\n- Prefer the user's own notes when they conflict with a source, and flag the disagreement.",
**kw
):
```
*A `rishi.Chat` bound to this vault’s system prompt, cached on the
vault.*
Four backends, one call: `litert` (CPU, no API key, the default because
it runs anywhere), `mlx` (Apple silicon), `llama` (any GGUF, needs
`rishi[llama]`) and `remote` (hosted, via fastllm). Pass `runtime=` only
when the model id cannot say for itself.
## Try it
Retrieval needs no model; only the answering step does.
[`mk_prompt`](https://vedicreader.github.io/vishalakshi/ask.html#mk_prompt)
is the whole contract, so it is worth looking at what the model actually
sees.
``` python
v = Vault(':memory:')
v.note('Late chunking beats naive chunking because context survives the split.')
print(mk_prompt('why late chunking?', v.context('late chunking'))[:400])
```
/Users/71293/code/personal/orgs/vishalakshi/.venv/lib/python3.13/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
from .autonotebook import tqdm as notebook_tqdm
[1] Late chunking beats naive chunking because context survives the split.
(source: note:4282b2c0f97d, pages 0–0)
Late chunking beats naive chunking because context survives the split.
---
Question: why late chunking?
/Users/71293/code/personal/orgs/vishalakshi/.venv/lib/python3.13/site-packages/litesearch/core.py:157: UserWarning: 'store': vector search using dtype='f16' but the store is registered as 'f32'. Distances are being computed over reinterpreted bytes — pass dtype= matching the embeddings you inserted.
warnings.warn(f'{tbl.name!r}: vector search using dtype={want!r} but the store is registered '
``` python
test_eq(resolve_model('mlx-community/Qwen3-4B-4bit'), ('mlx-community/Qwen3-4B-4bit', 'mlx'))
test_eq(resolve_model('my-local.gguf', 'llama'), ('my-local.gguf', 'llama'))
test_fail(lambda: resolve_model('gemma-3-4b-it-int4'), contains='backend')
```
``` python
res = L([AttrDict(node_id='d#1', title='A', breadcrumb='A › B', filename='f', doc_id='d')])
test_eq(cited('as [1] shows, and again [1], but not [9]', res).attrgot('node_id'), ['d#1'])
```
``` python
# reasoning arrives three ways: a full pair, an opener cut off at the cap, and — from an MLX chat
# template that prefills the opener — a bare closing tag with the reasoning in front of it
test_eq(split_reasoning('weighing [1]\n\nRRF fuses ranks [2].'),
('RRF fuses ranks [2].', 'weighing [1]'))
test_eq(split_reasoning('weighing [1]\n\nRRF fuses ranks [2].'),
('RRF fuses ranks [2].', 'weighing [1]'))
test_eq(split_reasoning('RRF fuses ranks [2].'), ('RRF fuses ranks [2].', ''))
# a section the reasoning weighed and dropped must not come back as a citation
test_eq(cited(split_reasoning('cites [1] cites nothing')[0], res), [])
```