tlg
3edc055299
- Add harmony.py: strip GPT-OSS-20B analysis/thinking channel from both streaming and non-streaming responses (HarmonyStreamFilter + extract_final_text) - Add per-model asyncio.Lock in llamacpp backend to prevent concurrent C++ access that caused container segfaults (exit 139) - Fix chat handler swap for streaming: move inside _stream_generate within lock scope (was broken by try/finally running before stream was consumed) - Filter /v1/models to return only LLM models (hide ASR/TTS from chat dropdown) - Correct Qwen3.5-4B estimated_vram_gb: 4 → 9 (actual allocation ~8GB) - Add GPU memory verification after eviction with retry loop in vram_manager - Add HF_TOKEN_PATH support in main.py for gated model access - Add /v1/audio/models and /v1/audio/voices discovery endpoints (no auth) - Add OOM error handling in both backends and chat route - Add AUDIO_STT_SUPPORTED_CONTENT_TYPES for webm/wav/mp3/ogg - Add performance test script (scripts/perf_test.py) - Update tests to match current config (42 tests pass) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
67 lines
2.0 KiB
Python
67 lines
2.0 KiB
Python
import pytest
|
|
|
|
from llmux.model_registry import ModelRegistry
|
|
|
|
|
|
@pytest.fixture
|
|
def registry():
|
|
return ModelRegistry.from_config()
|
|
|
|
|
|
def test_list_virtual_models(registry):
|
|
models = registry.list_virtual_models()
|
|
assert len(models) == 12 # only LLM models, not ASR/TTS
|
|
names = [m["id"] for m in models]
|
|
assert "Qwen3.5-9B-FP8-Thinking" in names
|
|
assert "GPT-OSS-20B-High" in names
|
|
assert "cohere-transcribe" not in names
|
|
assert "Chatterbox-Multilingual" not in names
|
|
|
|
|
|
def test_virtual_model_openai_format(registry):
|
|
models = registry.list_virtual_models()
|
|
m = next(m for m in models if m["id"] == "Qwen3.5-9B-FP8-Thinking")
|
|
assert m["object"] == "model"
|
|
assert m["owned_by"] == "llmux"
|
|
|
|
|
|
def test_resolve_virtual_to_physical(registry):
|
|
physical_id, physical, params = registry.resolve("Qwen3.5-9B-FP8-Thinking")
|
|
assert physical_id == "qwen3.5-9b-fp8"
|
|
assert physical.backend == "llamacpp"
|
|
assert params == {"enable_thinking": True}
|
|
|
|
|
|
def test_resolve_instruct_variant(registry):
|
|
physical_id, physical, params = registry.resolve("Qwen3.5-9B-FP8-Instruct")
|
|
assert physical_id == "qwen3.5-9b-fp8"
|
|
assert params == {"enable_thinking": False}
|
|
|
|
|
|
def test_resolve_gpt_oss_reasoning(registry):
|
|
physical_id, physical, params = registry.resolve("GPT-OSS-20B-Medium")
|
|
assert physical_id == "gpt-oss-20b"
|
|
assert params == {"system_prompt_prefix": "Reasoning: medium"}
|
|
|
|
|
|
def test_resolve_same_physical_for_variants(registry):
|
|
pid1, _, _ = registry.resolve("Qwen3.5-9B-FP8-Thinking")
|
|
pid2, _, _ = registry.resolve("Qwen3.5-9B-FP8-Instruct")
|
|
assert pid1 == pid2
|
|
|
|
|
|
def test_resolve_unknown_model_raises(registry):
|
|
with pytest.raises(KeyError):
|
|
registry.resolve("nonexistent-model")
|
|
|
|
|
|
def test_get_physical(registry):
|
|
physical = registry.get_physical("qwen3.5-9b-fp8")
|
|
assert physical.type == "llm"
|
|
assert physical.estimated_vram_gb == 10
|
|
|
|
|
|
def test_get_physical_unknown_raises(registry):
|
|
with pytest.raises(KeyError):
|
|
registry.get_physical("nonexistent")
|