typellm 0.1.0__tar.gz

typellm-0.1.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

typellm-0.1.0/PKG-INFO

@@ -0,0 +1,412 @@
+Metadata-Version: 2.4
+Name: typellm
+Version: 0.1.0
+Summary: Type-Safe Decoding for Autoregressive LLMs
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://typellm.ai/
+Project-URL: Repository, https://github.com/TypeLLM/TypeLLM
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: jinja2>=3.1
+Requires-Dist: tokenizers>=0.23
+Requires-Dist: transformers>=5.0
+Dynamic: license-file
+
+
+
+<div align="center">
+
+<img width="1500" height="500" alt="typellm-banner" src="https://github.com/user-attachments/assets/b1f2dbc6-21b7-4222-a0fb-dacfe1650797" />
+
+# TypeLLM: LLMs with type-safe generation
+
+<h4 align="center">
+  <a href="https://typellm.ai/">Homepage</a>&nbsp; • &nbsp;
+  <a href="https://typellm.ai/blog/introducing-typellm">Blog</a>&nbsp; • &nbsp;
+  <a href="https://typellm.ai/contact">Contact</a>&nbsp;
+</h4>
+</div>
+
+
+
+
+
+### Updates
+
+- **[2026/09/19]** Added optional [thinking mode](#thinking-mode) with
+  `thinking=True/False` and a configurable per-field thinking budget, followed
+  by type-safe constrained decoding. Thinking is off by default.
+- [2026/09/18] Added integer and float outputs through tokenizer-native
+  constrained decoding for JSON Schema `integer` and `number` fields.
+
+## Introduction
+
+TypeLLM extends autoregressive LLMs with type-safe generation. Models can still think and generate freely when needed, while producing guaranteed typed outputs when structure matters. Define the output with a JSON Schema, and TypeLLM returns values your software can use directly.
+
+### Supported output types
+
+- **Text** — Free text (`string`).
+- **Integer** — Whole numbers (`integer`).
+- **Number** — Numeric values (`number`).
+- **Boolean** — `true` or `false`.
+- **Enum choice** — One of your allowed string or numeric values.
+
+Enum and boolean fields select from finite candidates; numeric and text fields
+without `enum` generate values token by token. See [schemas and examples](#output-types).
+
+### Features
+
+1. **No out-of-schema hallucinations** — Choices stay within the allowed values.
+2. **Negligible output-token cost** — Single-token categorical selection and bounded numeric decoding; optional thinking adds tokens.
+3. **Linear input computation cost** — Prefix caching avoids reprocessing shared context.
+4. **Batch or sequential execution** — Run independent decisions together or condition on earlier results.
+5. **Made for open autoregressive LLMs** — Use compatible models you already serve with SGLang.
+6. **Supports thinking mode** — Enable reasoning before the final constrained answer.
+
+## Quick start
+
+### 1. Serve a model with SGLang
+
+Use [SGLang](https://github.com/sgl-project/sglang) to configure and serve a
+compatible autoregressive model on your local GPU server. This example uses
+Qwen3.8-27B; follow the
+[Qwen3.8-27B SGLang deployment guide](https://lmsysorg.mintlify.app/cookbook/autoregressive/Qwen/Qwen3.8-27B)
+to start it with prefix caching enabled.
+
+**Qwen3.5-4B and Qwen3.5-9B are also supported and GPU-tested, including thinking mode.**
+
+Serve the chosen checkpoint with SGLang and use the same model ID in the client:
+
+```python
+from typellm import TypeLLMClient
+
+client = TypeLLMClient(
+    "http://127.0.0.1:30000",
+    model="Qwen/Qwen3.8-27B",  # or "Qwen/Qwen3.5-4B", "Qwen/Qwen3.5-9B"
+)
+```
+
+Install the lightweight client-side tokenizer dependencies:
+
+```bash
+pip install -r requirements.txt
+```
+
+### 2. Run TypeLLM
+
+Point `TypeLLMClient` at the SGLang server's HTTP endpoint:
+
+```python
+from typellm import TypeLLMClient
+
+client = TypeLLMClient(
+    "http://127.0.0.1:30000",
+    model="Qwen/Qwen3.8-27B",
+)
+
+result = client.generate(
+    context="""
+    Receipt from Hilton London
+    Total: £324
+    Employee travelled to London for a client meeting.
+    """,
+    questions={
+        "expense_type": {
+            "type": "string",
+            "enum": ["meal", "travel", "equipment"],
+            "instructions": "What type of expense is this?",
+        },
+        "reimbursable": {
+            "type": "boolean",
+            "instructions": "Should this expense be reimbursed?",
+        },
+        "confidence": {
+            "type": "number",
+            "enum": [0.0, 0.25, 0.5, 0.75, 1.0],
+            "instructions": "How confident are you?",
+        },
+    },
+)
+
+print(result)
+# {
+#     "expense_type": "travel",
+#     "reimbursable": True,
+#     "confidence": 0.75,
+# }
+```
+
+`questions` maps output field names to their definitions. Every field is answered.
+The existing `schema=` JSON Schema interface is also supported; pass only one.
+`state=` is an alias for `context=`; pass only one of them.
+
+Python dictionary insertion order determines the decision order. Each later
+field is conditioned on the original context and the values selected for all
+earlier fields.
+
+## Thinking mode
+
+Thinking is off by default. Enable it when constructing the client:
+
+```python
+client = TypeLLMClient(
+    "http://127.0.0.1:30000",
+    model="Qwen/Qwen3.8-27B",
+    thinking=True,          # False disables thinking (the default)
+)
+result = client.generate(context=context, questions=questions)
+```
+
+No thinking-token budget is set by default. Optionally pass `thinking_budget=2048`
+to cap reasoning per field. TypeLLM reserves context space for the final answer;
+if thinking reaches its length limit, it keeps the partial reasoning, closes the
+thinking block, and proceeds with constrained decoding. Server and network errors
+still propagate.
+
+## Testing
+
+To run the 128-case numeric regression, first serve `Qwen/Qwen3.8-27B` with SGLang
+at `http://127.0.0.1:30000`, then run:
+
+```bash
+python3 _numeric_eval.py
+```
+
+The evaluation covers integer and number extraction, arithmetic, negative
+integers, and sequential dependencies. Its deterministic test cases are stored
+in `evals/numeric_eval_cases.jsonl`; the script writes detailed results to
+`evals/numeric_eval_formal_results.jsonl` and prints an aggregate summary.
+
+## Output types
+
+TypeLLM supports finite decisions, numeric fields, and free text:
+
+| Field | Schema | Returned value |
+|---|---|---|
+| Text | `{"type": "string"}` | `str` |
+| Integer | `{"type": "integer"}` | `int` |
+| Number | `{"type": "number"}` | `float` |
+| Boolean | `{"type": "boolean"}` | `bool` |
+| Enum choice | `{"type": "string", "enum": ["meal", "travel"]}` | Candidate type: `str`, `int`, or `float` |
+
+Enum choices support `string`, `integer`, and `number` types, with at most 16 values. The declared `type` validates the candidate values.
+
+Generation works in three ways:
+
+- **Choice** — Selects from finite candidates for enum and boolean fields.
+- **Numeric** — Generates integers or numbers without `enum` token by token under numeric constraints.
+- **Text** — Generates a JSON string for strings without `enum`, then decodes it to `str`.
+
+A string without `enum` generates free text:
+
+```python
+result = client.generate(
+    context="The train ticket is for a client meeting.",
+    questions={
+        "summary": {"type": "string", "instructions": "Summarize in one sentence."},
+    },
+)
+```
+
+`maxLength` is optional: add `"maxLength": 100` to limit Unicode character count.
+Omitting it adds no character limit. Set `TypeLLMClient(text_max_tokens=512)` to
+control the separate per-field generation budget (default 512 tokens).
+Incomplete, invalid, or over-length text raises `SGLangError`.
+Sequential fields can use earlier text; batch text fields generate independently.
+Text generation uses
+multiple tokens; type safety does not guarantee factual accuracy. Text fields
+currently support `maxLength`, but not `minLength`, `pattern`, or `format`.
+
+
+For example, ask for a numeric answer without enumerating every possible value:
+
+```python
+result = client.generate(
+    context="Calculate the requested value accurately.",
+    questions={
+        "answer": {
+            "type": "number",
+            "instructions": "What is 17.5 multiplied by 4?",
+        },
+    },
+)
+
+print(result)
+# {"answer": 70.0}
+```
+
+Use `instructions` to tell the model what decision to make:
+
+```python
+{
+    "type": "string",
+    "enum": ["billing", "technical", "account"],
+    "instructions": "Which team should handle this ticket?",
+}
+```
+
+If `instructions` is omitted, TypeLLM uses `description` or an instruction
+generated from the field name. Rename old `question` / `x-question` fields
+to `instructions`.
+
+## Sequential and batch execution
+
+Sequential execution is the default:
+
+Questions can express a complete decision workflow. For example, incident
+triage might select, in order:
+
+1. the affected system;
+2. the severity, conditioned on that system;
+3. whether to roll back, conditioned on both earlier decisions;
+4. a confidence score.
+
+Each field becomes a new user turn, and the assistant directly emits its
+single-token label or constrained numeric value. The completed turn is appended
+before the next question, so later decisions see the complete decision history.
+The final accumulated prompt is available as `client.last_prompt`, or can be
+printed with `print_final_prompt=True`.
+
+When the fields are independent, run them as one native SGLang batch:
+
+```python
+result = client.generate(
+    context=context,
+    questions=questions,
+    execution="batch",
+)
+```
+
+Batch execution prefills the shared context once, then forks it into one branch
+per field. Each branch appends only its own question and generates one token.
+The completed branch prompts are available as `client.last_prompts`.
+
+Use `sequential` when later decisions depend on earlier values. Use `batch`
+only when every field may be decided independently from the shared context.
+For actual concurrent execution, configure the SGLang server with
+`--max-running-requests` at least as large as the desired number of branches.
+
+### Batch performance
+
+Batch execution supports any number of independent fields, subject to the
+SGLang server's concurrency and memory limits. The shared context is prefilled
+once, and every field becomes a branch containing only its own question and
+one-token answer.
+
+As one illustrative measurement, a local run used Qwen3.8-27B NVFP4 on one
+NVIDIA RTX PRO 6000 Blackwell GPU, a roughly 1,100-token shared context, and
+`K=16` one-token Boolean decisions. SGLang was configured with
+`--max-running-requests 16`.
+
+| Execution | End-to-end latency | Latency per decision | Relative throughput |
+|---|---:|---:|---:|
+| Sequential | 9.35 s | 0.584 s | 1.0x |
+| Batch | 1.61 s | 0.101 s | 5.8x |
+
+In this run, every branch reused 1,088 cached tokens, for 17,408 reused token
+positions in total. This is a single example of the general batch method, not
+a fixed-width design or a portable hardware benchmark. Latency depends on
+`K`, the model, questions, context length, GPU, and server configuration.
+
+The two modes intentionally compute different conditionals. Sequential mode
+includes all earlier selected values in every later prompt. Batch mode gives
+each branch only the common context and its own question, which enables
+parallelism but removes cross-decision dependencies.
+
+## Probabilities and sampling
+
+Set `return_probabilities` on individual enum or boolean fields:
+
+```python
+result = client.generate(
+    context=context,
+    questions={
+        "expense_type": {
+            "type": "string",
+            "enum": ["meal", "travel", "equipment"],
+            "return_probabilities": True,
+        },
+    },
+)
+```
+
+```python
+{
+    "expense_type": {
+        "value": "travel",
+        "probabilities": {
+            "meal": 0.04,
+            "travel": 0.93,
+            "equipment": 0.03,
+        },
+    }
+}
+```
+
+Only opted-in fields return `value` and `probabilities`; other fields return plain values.
+The option is not supported on open Numeric or Text fields.
+
+Argmax is the default. To enable sampling:
+
+```python
+client = TypeLLMClient(
+    "http://127.0.0.1:30000",
+    mode="sample",
+    temperature=0.8,
+    seed=42,
+)
+```
+
+Sampling applies to finite candidates for Choice fields and to token generation
+for Numeric and Text fields. `temperature` controls sampling in each case.
+
+For a one-off request, use the convenience function:
+
+```python
+from typellm import run_schema
+
+result = run_schema(
+    context=context,
+    questions=questions,
+    base_url="http://127.0.0.1:30000",
+    model="Qwen/Qwen3.8-27B",
+)
+```
+
+## Cost analysis
+
+**Closed decisions generate one token per field, and prefix reuse makes the
+newly processed input grow approximately linearly with the unique context
+added across the workflow.** An open number requires one constrained step per
+generated tokenizer token. The long original context is normally prefilled once
+rather than recomputed for every decision.
+
+For `D` decisions, an original context of `C` tokens, and roughly `S` newly
+appended tokens per decision:
+
+```text
+without prefix reuse: O(D*C + D^2*S)
+with prefix reuse:    O(C + D*S)
+output generation:    O(D + N)
+```
+
+Here `N` is the total number of generated tokenizer tokens in open numeric
+fields, including their end-of-message tokens, and is zero for a fully finite
+schema.
+
+For `K` independent closed batch decisions with question lengths
+`Q_1, ..., Q_K`, the corresponding prefill count is approximately
+`C + sum(Q_k)`, followed by one batched decode step that produces `K` output
+tokens. Open numeric fields require additional constrained token steps.
+
+These are prefill token-position counts, not exact GPU FLOPs. New tokens still
+attend to the cached prefix, and real latency also depends on cache alignment,
+context length, batching, memory bandwidth, and cache eviction.
+
+This describes self-hosted compute. A hosted provider may still bill the full
+submitted input unless it offers cached-input pricing.
+
+## License
+
+[Apache License 2.0](LICENSE).