tenstorrent tt-inference-server issues

tenstorrent / tt-inference-server

Apache License 2.0

15 stars 2 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Extend trace region size for Llama TG

#321 ppetrovicTT opened 4 days ago
0
Fix TTQwen2ForCausalLM registration error with graceful fallback

#320 stisiTT closed 3 days ago
1
[Bug] vLLM Inference API Needs Triton

#319 ryanzhang-hlit closed 5 days ago
0
Qwen3-32B integration

#318 bgoelTT opened 5 days ago
0
quick try

#317 acvejicTT opened 6 days ago
0
Llama-3.3-70B benchmark on TG breaches trace region size

#316 ppetrovicTT opened 6 days ago
0
Update to hardcoded benchmark targets

#315 stisiTT opened 1 week ago
0
Max Sequence Length across high concurrency

#314 stisiTT opened 1 week ago
7
tstesco/fix-override-tt-config

#313 tstescoTT closed 1 week ago
0
Adding Image Benchmarks for Vision Language Models

#312 ssanjayTT closed 4 days ago
2
Add Eval implementation: LiveCodeBench

#311 tstescoTT opened 1 week ago
15
Fixing reference map bug

#310 ssanjayTT closed 1 week ago
3
Missing targets report

#309 acvejicTT closed 5 days ago
3
Missing dispatch_core_axis in override_tt_config on Llama-70B on TG

#308 ppetrovicTT closed 6 days ago
3
tstesco/benchmark-script-uplift

#307 tstescoTT closed 1 week ago
0
Rename LLAMA_DIR to CKPT_DIR

#306 rdraskicTT closed 1 week ago
3
Benchmarking should capture traces in ascending ISL to catch where issues are

#305 tstescoTT closed 6 days ago
0
integration with vLLM benchmark script with ISL template correction fix

#304 tstescoTT opened 1 week ago
0
Model Readiness Release: Mistral-7B-Instruct-v0.3

#303 ssanjayTT opened 1 week ago
3
vLLM rebase to upstream integration

#302 tstescoTT closed 1 week ago
2
Add detailed report

#301 acvejicTT closed 1 week ago
1
Update register_tt_models to use string paths instead of imports to support newer vLLM versions

#300 skhorasganiTT closed 1 week ago
2
Adding benchmark support for Mistral-7B-Instruct

#299 ssanjayTT closed 1 week ago
0
Faster hang / crash detection in evals and benchmark client side scripts

#298 tstescoTT opened 2 weeks ago
0
tstesco/model-spec-params

#297 tstescoTT closed 1 week ago
1
fix passing only staged changed files to ruff

#296 tstescoTT closed 2 weeks ago
0
adding tt-metal build within container user to avoid many file permissions changes

#295 tstescoTT closed 2 weeks ago
1
tstesco/cleanup-files

#294 tstescoTT closed 2 weeks ago
0
add "$@" arg .pre-commit-config.yaml

#293 tstescoTT closed 2 weeks ago
0
fix pre-commit to only check modified files

#292 tstescoTT closed 2 weeks ago
0
tstesco/fix-pre-commit-ruff

#291 tstescoTT closed 2 weeks ago
0
Make sure to install torch for cpu

#290 ppetrovicTT closed 2 weeks ago
0
Extend trace region size for Llama70B on TG

#289 ppetrovicTT closed 2 weeks ago
0
Set timeout params in model_config.py to reduce CI runtimes on models that have failed to start

#288 tstescoTT opened 2 weeks ago
0
Image-text benchmarks for multimodal models

#287 ssanjayTT closed 3 days ago
1
Failures handling

#286 ppetrovicTT closed 2 weeks ago
0
Whisper-Distil-Large-v3 performance benchmarks

#285 bgoelTT opened 2 weeks ago
0
Whisper-Distil-Large-v3 accuracy evaluations

#284 bgoelTT opened 2 weeks ago
0
tstesco/authless-client-scripts

#283 tstescoTT closed 2 weeks ago
2
Use product names for all printouts and generated reports

#282 tstescoTT opened 2 weeks ago
0
tstesco/example-model-cli-input

#281 tstescoTT closed 2 weeks ago
0
Enable client side scripts for benchmarking, evals, and tests to run against vLLM example server

#280 tstescoTT closed 2 weeks ago
1
tstesco/example script no auth

#279 tstescoTT closed 2 weeks ago
0
fix link to development docs

#278 anirudTT closed 2 weeks ago
0
Benchmarks trying unsupported sequence length

#277 ppetrovicTT closed 1 week ago
7
Qwen2.5-VL-72B shows garbage output tokens in server mode after serving the first prompt

#276 gwangTT opened 2 weeks ago
1
Accuracy evaluations: Llama-3.1-8B on Blackhole

#275 bgoelTT opened 2 weeks ago
0
Performance benchmarks: Llama-3.1-8B on Blackhole

#274 bgoelTT opened 2 weeks ago
0
Model readiness support: Llama-3.1-8B on Blackhole

#273 bgoelTT opened 2 weeks ago
0
Enabling Server-side Tokenization, Prompt Generation, and Streamlined Benchmarking

#272 stisiTT closed 1 week ago
10