Conversion of RawTextItem to TextContentItem causes no tokens to be generated

[AidanFRyan]

System Info

/home/aidan/miniconda3/envs/llamastack-meta-reference-gpu/lib/python3.10/runpy.py:126: RuntimeWarning: 'torch.utils.collect_env' found in sys.modules after import of package 'torch.utils', but prior to execution of 'torch.utils.collect_env'; this may result in unpredictable behaviour
warn(RuntimeWarning(msg))
Collecting environment information...
PyTorch version: 2.5.1+cu124
Is debug build: False
CUDA used to build PyTorch: 12.4
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.3 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.22.1
Libc version: glibc-2.35

Python version: 3.10.16 (main, Dec 11 2024, 16:24:50) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-107-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.6.85
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA GeForce RTX 3090
Nvidia driver version: 565.77
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 43 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: AuthenticAMD
Model name: AMD Ryzen 7 3800X 8-Core Processor
CPU family: 23
Model: 113
Thread(s) per core: 2
Core(s) per socket: 8
Socket(s): 1
Stepping: 0
Frequency boost: enabled
CPU max MHz: 4500.0000
CPU min MHz: 2200.0000
BogoMIPS: 9000.03
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca sme sev sev_es
Virtualization: AMD-V
L1d cache: 256 KiB (8 instances)
L1i cache: 256 KiB (8 instances)
L2 cache: 4 MiB (8 instances)
L3 cache: 32 MiB (2 instances)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Retbleed: Mitigation; untrained return thunk; SMT enabled with STIBP protection
Vulnerability Spec rstack overflow: Mitigation; safe RET
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Retpolines; IBPB conditional; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected

Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] torch==2.5.1
[pip3] torchvision==0.20.1
[pip3] triton==3.1.0
[conda] torch 2.5.1 pypi_0 pypi
[conda] torchvision 0.20.1 pypi_0 pypi
[conda] triton 3.1.0 pypi_0 pypi

Information

• The official example scripts
• My own modified scripts

🐛 Describe the bug

Receive pydantic warnings at call of llama_stack/providers/inline/inference/meta_reference/inference.py:299, debug shows RawTextItem content type for each message passed in _nonstream_chat_completion converts to TextContentItem at llama_stack/providers/inline/inference/meta_reference/generation.py:381, which then prevents the isinstance(c, RawTextItem) at llama_models/llama3/api/chat_format.py:79 from tokenizing the message content when using meta-reference-gpu.

My temporary fix is converting message content to RawTextItem in llama_stack/providers/inline/inference/meta_reference/generation.py:382 as follows:

for message in request.messages:
if isinstance(message.content, TextContentItem):
message.content = RawTextItem(text=message.content.text)

This fix requires the relevant import statements.

Error logs

Warning in output:

/miniconda3/envs/llamastack-meta-reference-gpu/lib/python3.10/site-packages/pydantic/main.py:477: UserWarning: Pydantic serializer warnings:
PydanticSerializationUnexpectedValue: Expected ReadyRequest but got TaskRequest with value TaskRequest(type=<Process...m=False, logprobs=None)) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected ReadyResponse but got TaskRequest with value TaskRequest(type=<Process...m=False, logprobs=None)) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected EndSentinel but got TaskRequest with value TaskRequest(type=<Process...m=False, logprobs=None)) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected CancelSentinel but got TaskRequest with value TaskRequest(type=<Process...m=False, logprobs=None)) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: PydanticSerializationUnexpectedValue: Expected CompletionRequest but got ChatCompletionRequest with value ChatCompletionRequest(mod...am=False, logprobs=None) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: PydanticSerializationUnexpectedValue: Expected UserMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected SystemMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected ToolResponseMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected CompletionMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected TaskResponse but got TaskRequest with value TaskRequest(type=<Process...m=False, logprobs=None)) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected ExceptionResponse but got TaskRequest with value TaskRequest(type=<Process...m=False, logprobs=None)) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected CompletionRequest but got ChatCompletionRequest with value ChatCompletionRequest(mod...am=False, logprobs=None) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: PydanticSerializationUnexpectedValue: Expected UserMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected SystemMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected ToolResponseMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected CompletionMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected UserMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected SystemMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected ToolResponseMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected CompletionMessage but got RawMessage with value RawMessage(role='system',...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected UserMessage but got RawMessage with value RawMessage(role='user', c...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected SystemMessage but got RawMessage with value RawMessage(role='user', c...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected ToolResponseMessage but got RawMessage with value RawMessage(role='user', c...son=None, tool_calls=[]) - serialized value may not be as expected
PydanticSerializationUnexpectedValue: Expected CompletionMessage but got RawMessage with value RawMessage(role='user', c...son=None, tool_calls=[]) - serialized value may not be as expected
return self.pydantic_serializer.to_json(

Expected behavior

Should pass RawTextItem content type to Llama::chat_completion() at generation.py:381, unsure why this conversion takes place.