tensorzero tensorzero issues

tensorzero / tensorzero

make LLMs improve through experience

https://tensorzero.com

Apache License 2.0

395 stars 16 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Allow text blocks alongside tool calls in provider tests

#429 virajmehta closed 23 hours ago
0
Tests are failing more than is needed when models return text + tool call

#428 virajmehta closed 23 hours ago
0
Consider how to track and limit usage for multiple users and service accounts

#427 GabrielBianconi opened 1 day ago
0
added uuid7 to client code

#426 virajmehta closed 22 hours ago
0
Add Typescript bindings for minijinja

#425 virajmehta opened 1 day ago
0
added Zod schemas for TensorZero gateway configuration

#424 virajmehta opened 2 days ago
1
Proposal: Implement an OpenAI-compatible endpoint for the gateway

#423 virajmehta opened 2 days ago
1
Consider how to integrate batch inference

#422 GabrielBianconi closed 2 days ago
1
Consider including a UUID7 implementation in the `tensorzero` Python package

#420 GabrielBianconi closed 22 hours ago
0
Build a Zod schema that matches the config tree

#419 virajmehta opened 4 days ago
0
Fixed issue in python client where exception handling failed due to no body in response

#418 virajmehta closed 4 days ago
0
Sonnet is throwing errors on GCP

#417 virajmehta closed 4 days ago
2
Explore how to include additional debugging information in inference provider-side errors

#416 virajmehta opened 5 days ago
1
Async Python client causes exception when handling 404 in a streaming request

#415 GabrielBianconi closed 4 days ago
0
Bump Python client version

#414 GabrielBianconi closed 5 days ago
0
Clickhouse typescript integration

#413 virajmehta closed 22 hours ago
0
Support OIDC authentication?

#412 cboettig opened 6 days ago
5
Trying to use with Ollama

#411 jeremyandrews closed 6 days ago
3
Warn if variant includes inference params not supported by model/providers

#410 GabrielBianconi opened 1 week ago
0
Add `top_p`, `presence_penalty`, and `frequency_penalty` options to inference requests

#409 virajmehta closed 1 week ago
1
Beautify charts and update notebooks

#408 eyadashrafkh opened 1 week ago
0
Add a recipe for fine-tuning on demonstrations using Fireworks

#407 virajmehta closed 22 hours ago
1
Consider how to include billing information for observability

#406 GabrielBianconi opened 1 week ago
0
Update README

#405 GabrielBianconi closed 2 weeks ago
0
Migrate JSON columns to the ClickHouse JSON data type as appropriate

#404 virajmehta opened 2 weeks ago
1
Consider adding an integration with TAU-bench

#403 virajmehta opened 2 weeks ago
0
fixed python client to do streaming inference properly

#402 virajmehta closed 2 weeks ago
0
Add a version endpoint to gateway

#401 virajmehta opened 2 weeks ago
0
Add "hydrated" inputs and outputs to ClickHouse ModelInference table

#400 virajmehta closed 1 week ago
0
Consider the correct data model for embedding ModelInferenceRequests as we move towards an embedding gateway

#399 virajmehta opened 2 weeks ago
0
Improve documentation on best-of-n and mixture-of-n examples to explain what kind of prompts need to be written

#398 virajmehta opened 2 weeks ago
0
Add a production deployment example and also CI tests that check that it works

#397 virajmehta closed 1 week ago
0
Add Mixture-of-N Sampling inference-time optimization (variant type)

#396 virajmehta closed 2 weeks ago
0
Implemented tagging for `/feedback` endpoint

#395 virajmehta closed 2 weeks ago
0
Implemented migrations for tagged feedback

#394 virajmehta closed 2 weeks ago
0
Implement a mixture-of-n / fuser type variant

#393 virajmehta closed 2 weeks ago
0
Implement prefilling for JSON mode for Anthropic models

#392 virajmehta closed 3 weeks ago
0
Update Fireworks fine-tuning API to use the REST interface

#391 virajmehta opened 3 weeks ago
0
Consider ways to store `logprobs`

#389 GabrielBianconi opened 3 weeks ago
0
Implement configurable retries at Variant level

#388 virajmehta closed 3 weeks ago
0
Add a feature that says "retry if JSON is noncompliant with output_schema" to JSON type functions

#387 virajmehta opened 3 weeks ago
0
Integrated Google AI Studio Gemini provider

#385 virajmehta closed 3 weeks ago
1
Test Ollama integration

#384 GabrielBianconi opened 3 weeks ago
1
implement Anthropic `disable_parallel_tools` flag

#383 virajmehta opened 4 weeks ago
0
Anthropic now allows multiple consecutive user / assistant messages, remove special handling for that.

#382 virajmehta opened 4 weeks ago
0
Update README.md

#381 GabrielBianconi closed 3 weeks ago
0
Update README.md

#380 GabrielBianconi closed 4 weeks ago
0
Think about how to best support batched inferences with e.g. anthropic and OpenAI async APIs

#379 virajmehta closed 3 weeks ago
1
Make sure prompt caching is implemented as-needed for each of the many inference providers that support it

#378 virajmehta opened 4 weeks ago
0
Make it as easy as possible to create demonstrations for every time an LLM fails JSON schema validation

#377 virajmehta opened 4 weeks ago
0