TensorSharp
A native .NET LLM inference engine for GGUF models. TensorSharp provides a console application, a web-based chatbot interface, and Ollama/OpenAI-compatible HTTP APIs for programmatic access. It supports Windows/MacOS/Linux with full GPU capability
File Explorer
Download Latest Version (.zip)- deploy-pages.yml
- pr-unit-tests.yml
- publish-nuget.yml
- release-binaries.yml
- test-matrix.yml
- README.md
- AdvUtils.csproj
- Logger.cs
- ConcurrentDecodeBench.csproj
- Program.cs
- CpuQuantBench.csproj
- Program.cs
- CudaQuantizedMatmulBench.csproj
- Program.cs
- DecodeBench.csproj
- Program.cs
- config.cpython-310.pyc
- engines.cpython-310.pyc
- scenarios.cpython-310.pyc
- weather.json
- long_text.txt
- prefill_corpus.txt
- benchmark_config.json
- benchmark_config_ci.json
- benchmark_config_multigpu.json
- benchmark_config_prefill.json
- config.py
- download_models.py
- downloads.py
- engines.py
- README.md
- report.py
- run_matrix.py
- scenarios.py
- FlashAttnBench.csproj
- Program.cs
- GdnDecodeBench.csproj
- Program.cs
- PrefillBench.csproj
- Program.cs
- Program.cs
- QwenImageEditBench.csproj
- Program.cs
- SafetensorsLoadBench.csproj
- Program.cs
- VisionEncoderBench.csproj
- Program.cs
- WanVideoBench.csproj
- auto-download.json
- cli-basic.json
- diffusiongemma-26b-a4b-q3.json
- diffusiongemma-26b-a4b-q4.json
- gemma-4-12b.json
- gemma-4-26b-a4b.json
- gemma-4-e4b.json
- gpt-oss-20b.json
- qwen-image-edit-2511.json
- qwen-image-edit.json
- qwen-image-rapid-nsfw.json
- qwen3.5-9b-iq4_xs.json
- qwen3.5-9b-q8.json
- qwen3.5-9b-uncensored-q8.json
- qwen3.6-27b.json
- qwen3.6-35b-a3b.json
- README.md
- server-basic.json
- variables.json
- deepseek4.md
- deepseek4_zh-cn.md
- diffusiongemma.md
- diffusiongemma_zh-cn.md
- gemma3.md
- gemma3_zh-cn.md
- gemma4.md
- gemma4_zh-cn.md
- glm.md
- glm_zh-cn.md
- gptoss.md
- gptoss_zh-cn.md
- mistral3.md
- mistral3_zh-cn.md
- muse-glimmer.md
- muse-glimmer_zh-cn.md
- nemotron.md
- nemotron_zh-cn.md
- qwen3.md
- qwen35.md
- qwen35_zh-cn.md
- qwen3_zh-cn.md
- qwenimage.md
- qwenimage_zh-cn.md
- README.md
- README_zh-cn.md
- wan.md
- wan_zh-cn.md
- gemma4-prefill-cuda-graph-design.md
- BOOK.md
- BOOK_zh-cn.md
- engine_comparison_report.md
- env_var_feature_matrix.md
- env_var_feature_matrix_zh-cn.md
- model_cards.md
- model_cards_zh-cn.md
- moe_cpu_offload_benchmark.md
- PAGED_ATTENTION_AND_CONTINUOUS_BATCHING.md
- PAGED_ATTENTION_AND_CONTINUOUS_BATCHING_zh-cn.md
- structured_output_grammar.md
- dsv4-dspark-to-gguf.py
- ensure-metal-toolchain.sh
- fetch-ggml.ps1
- fetch-ggml.sh
- fetch-vulkan-toolchain.ps1
- fetch-vulkan-toolchain.sh
- make-test-media.sh
- verify-packages.ps1
- vs-locate.ps1
- assistant_logo.png
- assistant_logo_2.png
- assistant_logo_3.png
- banner_1.png
- wan22_oracle.py
- ApiExceptionMiddlewareTests.cs
- ApiModelsResponseTests.cs
- BackendCatalogTests.cs
- BannedSymbols.txt
- BatchedExecutorTests.cs
- BpeTokenizerTests.cs
- ChatAttachmentPathGuardTests.cs
- ChatGenerationPipelineDocumentTests.cs
- ChatSessionTests.cs
- ClampGenerationReserveTests.cs
- ConfigFileArgsTests.cs
- ContinuousBatchSchedulerTests.cs
- CpuKernelFastPathTests.cs
- CudaAllocatorConcurrencyTests.cs
- CudaBackendTests.cs
- CudaDecodeAttentionWrapTests.cs
- CudaGraphPocTests.cs
- CudaKvCacheResizeTests.cs
- CudaLongDecodeReproTests.cs
- CudaQ8MatmulHarnessTests.cs
- DecodeBackendParityTests.cs
- DeepSeek4ToolCallTests.cs
- DiffusionGemmaTests.cs
- ElementwiseRowBroadcastTests.cs
- EngineParallelInferenceTests.cs
- EnvScope.cs
- ExecutionPlannerTests.cs
- FileLoggerProviderTests.cs
- FinishReasonMappingTests.cs
- GatedFacts.cs
- GatherScatterTests.cs
- Gemma4AudioEncoderTests.cs
- Gemma4BatchedForwardTests.cs
- Gemma4BatchedPerfBench.cs
- Gemma4MtpTests.cs
- Gemma4PrimedThinkingParserTests.cs
- Gemma4PromptRenderReproTests.cs
- GgmlAttentionTests.cs
- GgmlBackendTestInitializer.cs
- GgmlCopyDtypeTests.cs
- GgmlCopyStridedFloat32Tests.cs
- GgmlMuseGlimmerVisionBlockTests.cs
- GlmDsaMtpModelTests.cs
- GlmDsaMtpTests.cs
- GlmDsaParityTests.cs
- GlmDsaSyntheticModelBuilder.cs
- GlmDsaTinyModelTests.cs
- GlobalUsings.cs
- GptOssBatchedCorrectnessTests.cs
- GptOssBatchedPerfBench.cs
- GptOssLongDecodeReproTests.cs
- GrammarConstrainedDecodingTests.cs
- HarmonyToolCallIntegrationTests.cs
- HarmonyToolCallTests.cs
- ImageProcessorTests.cs
- InferenceWeb.Tests.csproj
- Jinja2TemplateTests.cs
- KvBlockTransferRingTests.cs
- KvCacheDtypeQ4Tests.cs
- KVCacheMultiTurnIntegrationTests.cs
- KVCachePromptRendererTests.cs
- KVCacheTests.cs
- ManagedQuantizedOpsTests.cs
- MediaHelperTests.cs
- Mistral3BatchedForwardTests.cs
- MlxBackendTests.cs
- MlxIq3XxsKernelTests.cs
- MlxIQuantDecodeMatmulTests.cs
- MlxStorageBulkElementIoTests.cs
- ModelContextLengthTests.cs
- ModelMultimodalInjectorCoverageTests.cs
- ModelReloadRollbackTests.cs
- ModelServiceHistoryTests.cs
- ModelServiceRawTokenHistoryTests.cs
- ModelServiceSessionTests.cs
- ModelServiceUploadLoggingTests.cs
- MoeCpuOffloadConfigTests.cs
- MoeCpuOffloadDescriptorLayoutTests.cs
- MtpSpeculativeExecutionTests.cs
- MuseGlimmerKvReuseRenderTests.cs
- MuseGlimmerKvSnapshotTests.cs
- MuseGlimmerMlxDecodeTests.cs
- MuseGlimmerOutputParserTests.cs
- MuseGlimmerParityTests.cs
- MuseGlimmerRegressionTests.cs
- NemotronBatchedCorrectnessTests.cs
- NemotronBatchedPerfBench.cs
- NemotronConvTests.cs
- PagedKvCacheCodecTests.cs
- PagedKvCacheTests.cs
- ParallelThroughputBench.cs
- PdfPageImageExtractorTests.cs
- PdfTextExtractorTests.cs
- PrefillChunkingTests.cs
- PrefillOptimizationBenchmark.cs
- PromptOverflowMiddlewareTests.cs
- Qwen35BatchedCorrectnessTests.cs
- Qwen35BatchedPerfBench.cs
- Qwen35TokenDecodeContractTests.cs
- Qwen35ToolCallTests.cs
- Qwen36MtpTests.cs
- Qwen3BatchedForwardTests.cs
- QwenImageDiTWeightDtypeTests.cs
- QwenImageMultiImageTests.cs
- QwenImageQualityDefaultsTests.cs
- ReasoningContentStreamingTests.cs
- RedisCacheTests.cs
- RequestLoggingMiddlewareTests.cs
- ResponsesApiTests.cs
- RetainedFusedCacheTests.cs
- SafetensorsReaderTests.cs
- SamplingConfigCloneTests.cs
- SamplingConfigParserTests.cs
- SamplingPrecedenceTests.cs
- ServerOptionsBuilderTests.cs
- SessionManagerTests.cs
- StructuredOutputTests.cs
- TcpCommunicatorTests.cs
- TensorApiTests.cs
- TestAssemblyConfig.cs
- TextUploadHelperTests.cs
- TokenSamplerTests.cs
- ToolFunctionParserBenchmark.cs
- ToolFunctionParserTests.cs
- ToolFunctionParseTests.cs
- TurboQuantKvCodecBenchmark.cs
- TurboQuantKvCodecTests.cs
- UdQuantShardConfigTests.cs
- UploadContentPolicyTests.cs
- UploadFileReferenceTests.cs
- UploadRootConfinementTests.cs
- UploadStoragePolicyTests.cs
- VisionEncoderCudaTests.cs
- WanDirectOpsTests.cs
- WanDitBackendParityTests.cs
- WanVaeBackendParityTests.cs
- WanVideoOracleTests.cs
- WanVideoParamsParserTests.cs
- WanVideoTests.cs
- WebUiChatPolicyTests.cs
- Dsv4CudaEngine.cs
- Dsv4CudaEngine.Dspark.cs
- Dsv4CudaEngine.HostMoe.cs
- Dsv4Kernels.cs
- IDsv4HostMatMul.cs
- IDsv4WeightSource.cs
- CublasApi.cs
- CudaDriverApi.cs
- CudaErrorHelper.cs
- CudaException.cs
- CudaLibraryResolver.cs
- tensorsharp_dsv4_kernels.cu
- tensorsharp_dsv4_tables.cuh
- tensorsharp_iq2xxs_tables.cuh
- tensorsharp_kernels.cu
- tensorsharp_dsv4_kernels.ptx
- tensorsharp_kernels.ptx
- resolve-cuda-arch.ps1
- resolve-cuda-arch.sh
- AssemblyInfo.cs
- CudaAllocator.cs
- CudaBackend.cs
- CudaBasicOps.cs
- CudaBlas.cs
- CudaContext.cs
- CudaCpuFallback.cs
- CudaCublasHandle.cs
- CudaDevice.cs
- CudaDeviceMemoryPool.cs
- CudaEvent.cs
- CudaFusedOps.cs
- CudaGraphCapture.cs
- CudaKernelOps.cs
- CudaKernels.cs
- CudaModule.cs
- CudaP2PCommunicator.cs
- CudaProfileCounters.cs
- CudaQuantizedOps.cs
- CudaStorage.cs
- CudaStream.cs
- CudaWanOps.cs
- TensorParallelGroup.cs
- TensorSharp.Backends.Cuda.csproj
- GgmlAllocator.cs
- GgmlBasicOps.cs
- GgmlContext.cs
- GgmlDeepSeek4Native.cs
- GgmlGgufTensorDequant.cs
- GgmlGlmNative.cs
- GgmlLossOps.cs
- GgmlMemoryPool.cs
- GgmlNative.cs
- GgmlStorage.cs
- GgmlTensorParallel.cs
- GgmlTensorParallelGroup.cs
- TensorSharp.Backends.GGML.csproj
- CMakeLists.txt
- MLX_C_VERSION
- build-native-macos.sh
- MlxAllocator.cs
- MlxBackend.cs
- MlxBasicOps.cs
- MlxCompiledOps.cs
- MlxCpuFallback.cs
- MlxFallbackOps.cs
- MlxFusedOps.cs
- MlxMemorySnapshot.cs
- MlxNative.cs
- MlxQuantizedOps.cs
- MlxStorage.cs
- MlxWorker.cs
- MoeExpertOffload.cs
- TensorSharp.Backends.MLX.csproj
- CliLoggingSetup.cs
- batch_thinking.jsonl
- example_api_thinking_tools.md
- example_api_thinking_tools_zh-cn.md
- input_thinking.txt
- input_tool_calc.txt
- input_tool_call.txt
- kv_cache_smoke.jsonl
- kv_cache_thinking.jsonl
- tools_calculator.json
- tools_weather.json
- CliUsage.cs
- GlobalUsings.cs
- InteractiveSession.cs
- Program.cs
- TensorSharp.Cli.csproj
- test_requests.jsonl
- DelegateDisposable.cs
- TensorConcatenation.cs
- TensorResultBuilder.cs
- DGEMM.cs
- LSAME.cs
- SGEMM.cs
- XERBLA.cs
- CpuAllocator.cs
- CpuBasicOps.cs
- CpuFillCopyOps.cs
- CpuIndexingOps.cs
- CpuMaxPoolingOps.cs
- CpuNativeHelpers.cs
- CpuOpsNative.cs
- CpuRandom.cs
- CpuStorage.cs
- MatrixMultiplication.cs
- NativeWrapper.cs
- OpenBlasNative.cs
- SpatialConvolutionMM.cs
- SExpression.cs
- SVar.cs
- TExpression.cs
- TVar.cs
- FolderProfile.pubxml
- AssemblyInfo.cs
- DType.cs
- Half.cs
- IAllocator.cs
- IBasicOps.cs
- ITensorParallelGroup.cs
- OpConstraint.cs
- OpRegistry.cs
- OpRegistryAttributes.cs
- Ops.cs
- RandomGenerator.cs
- RefCounted.cs
- ReflectionExtensions.cs
- Storage.cs
- Tensor.cs
- TensorApplyCPU.cs
- TensorComputePrimitives.cs
- TensorDimensionHelpers.cs
- TensorDimIterState.cs
- TensorFormatting.cs
- TensorIterState.cs
- TensorSerialization.cs
- TensorShape.cs
- TensorSharp.Core.csproj
- TensorView.cs
- DistributedTensorParallelGroup.cs
- DistributedTpConfig.cs
- TcpCommunicator.cs
- TensorSharp.Distributed.csproj
- InferenceEngineSmoke.csproj
- Program.cs
- addmm_vector_test.cpp
- get_rows_quant_test.cpp
- build-linux.sh
- build-macos.sh
- build-windows.ps1
- CMakeLists.txt
- ggml_api_exports.def
- ggml_ops_core.cpp
- ggml_ops_deepseek4.cpp
- ggml_ops_dflash.cpp
- ggml_ops_diffusion.cpp
- ggml_ops_diffusion_sample.cu
- ggml_ops_dsv4_fused.cu
- ggml_ops_dsv4_fused.h
- ggml_ops_dsv4_fused_cpu.cpp
- ggml_ops_elementwise.cpp
- ggml_ops_fused.cpp
- ggml_ops_gated_delta_net.cpp
- ggml_ops_gemma4_batched.cpp
- ggml_ops_gemma4_decode.cpp
- ggml_ops_gemma4_moe.cpp
- ggml_ops_gemma4_verify.cpp
- ggml_ops_glm_dsa.cpp
- ggml_ops_gptoss_decode.cpp
- ggml_ops_gptoss_kv.h
- ggml_ops_gptoss_prefill.cpp
- ggml_ops_host_pin.cu
- ggml_ops_internal.h
- ggml_ops_mamba2.cpp
- ggml_ops_mask.cu
- ggml_ops_matmul.cpp
- ggml_ops_moe.cpp
- ggml_ops_muse_glimmer.cpp
- ggml_ops_muse_glimmer_vision.cpp
- ggml_ops_norm_attn.cpp
- ggml_ops_paged_attention.cpp
- ggml_ops_qwen35_decode.cpp
- ggml_ops_qwen35_gdn_tp.cpp
- ggml_ops_qwen35_recurrent_prefill.cpp
- ggml_ops_qwen35_verify.cpp
- ggml_ops_qwen_image.cpp
- ggml_ops_tensor_parallel.cpp
- ggml_ops_tp_probe.cu
- ggml_ops_training.cpp
- ggml_ops_transformer.cpp
- ggml_ops_transformer_common.h
- ggml_ops_transformer_prefill.cpp
- ggml_ops_wan.cpp
- tsg_cuda_cudnn_conv.cu
- tsg_metal_mps_conv.mm
- tsg_metal_msl_default.m
- DeepSeek4CpuExecutor.Attention.cs
- DeepSeek4CpuExecutor.cs
- DeepSeek4CudaExecutor.cs
- DeepSeek4Model.cs
- DeepSeek4Model.Dspark.cs
- DeepSeek4Model.PerSeqCache.cs
- Dsv4SpinPool.cs
- DiffusionGemmaModel.cs
- DiffusionGemmaSampler.cs
- Gemma3Model.cs
- Gemma3Model.TensorParallel.cs
- Gemma3VisionEncoder.cs
- ImageProcessor.cs
- Gemma4AudioEncoder.cs
- Gemma4AudioPreprocessor.cs
- Gemma4ImageProcessor.cs
- Gemma4Model.BatchedForward.cs
- Gemma4Model.cs
- Gemma4Model.CudaMoEPrefill.cs
- Gemma4Model.Mtp.cs
- Gemma4Model.PerSeqCache.cs
- Gemma4Model.TensorParallel.cs
- Gemma4Model.TensorParallelFusedDecode.cs
- Gemma4Model.TensorParallelMoEFused.cs
- Gemma4VisionEncoder.cs
- GlmDsaModel.Attention.cs
- GlmDsaModel.cs
- GlmDsaModel.Debug.cs
- GlmDsaModel.Moe.cs
- GlmDsaModel.Mtp.cs
- GlmDsaModel.MtpNative.cs
- GlmDsaModel.Native.cs
- GlmDsaModel.PerSeqCache.cs
- GlmDsaModel.TensorParallel.cs
- GptOssModel.BatchedForward.cs
- GptOssModel.cs
- GptOssModel.FusedModelDecode.cs
- GptOssModel.TensorParallel.cs
- GptOssModel.TensorParallelGgmlMoE.cs
- Mistral3ImageProcessor.cs
- Mistral3Model.BatchedForward.cs
- Mistral3Model.BatchedForwardTP.cs
- Mistral3Model.cs
- Mistral3Model.TensorParallel.cs
- Mistral3VisionEncoder.cs
- DFlashConfig.cs
- MuseGlimmerImageProcessor.cs
- MuseGlimmerModel.cs
- MuseGlimmerModel.DFlash.cs
- MuseGlimmerModel.DFlash.Fused.cs
- MuseGlimmerModel.Fused.cs
- MuseGlimmerModel.MlxDecode.cs
- MuseGlimmerModel.TensorParallel.cs
- MuseGlimmerModel.TensorParallelFused.cs
- MuseGlimmerVisionEncoder.cs
- NemotronAudioPreprocessor.cs
- NemotronImageProcessor.cs
- NemotronModel.BatchedForward.cs
- NemotronModel.cs
- NemotronModel.TensorParallel.cs
- NemotronVisionEncoder.cs
- Qwen3Model.BatchedForward.cs
- Qwen3Model.BatchedForwardTP.cs
- Qwen3Model.cs
- Qwen3Model.TensorParallel.cs
- ImageProcessor.cs
- Qwen35Model.BatchedForward.cs
- Qwen35Model.BatchedFusedDecode.cs
- Qwen35Model.cs
- Qwen35Model.CudaMoE.cs
- Qwen35Model.CudaMoEPrefill.cs
- Qwen35Model.GatedDeltaNet.cs
- Qwen35Model.Mtp.cs
- Qwen35Model.PerSeqCache.cs
- Qwen35Model.TensorParallel.cs
- Qwen35Model.TensorParallelFusedAttention.cs
- Qwen35Model.TensorParallelFusedModelDecode.cs
- Qwen35Model.TensorParallelGgmlMoE.cs
- Qwen35Model.TensorParallelMoE.cs
- Qwen35VisionEncoder.cs
- ImageIO.cs
- QwenImageDiT.Cache.cs
- QwenImageDiT.CfgBatch.cs
- QwenImageDiT.cs
- QwenImageDiT.Native.cs
- QwenImageDiT.WholeModel.cs
- QwenImageLora.cs
- QwenImageModel.cs
- QwenImageParams.cs
- QwenImagePipeline.cs
- QwenImagePrompt.cs
- QwenImageScheduler.cs
- QwenImageStepCache.cs
- QwenImageTextEncoder.cs
- QwenImageVae.cs
- QwenImageVae.Reference.cs
- QwenImageVaeGraph.cs
- QwenImageVisionEncoder.cs
- QwenImageVisionProcessor.cs
- VaeReferenceMath.cs
- VaeWeights.cs
- VideoIO.cs
- WanDirectDiT.cs
- WanDirectOps.cs
- WanDirectT5.cs
- WanDirectVae.cs
- WanDiT.cs
- WanNetworks.cs
- WanScheduler.cs
- WanTextEncoder.cs
- WanUniPCScheduler.cs
- WanVae.cs
- WanVaeBase.cs
- WanVaeWeights.cs
- WanVideoModel.cs
- WanVideoParams.cs
- WanVideoPipeline.cs
- TensorPagedAttention.cs
- BackendExecutionPlan.cs
- GlobalUsings.cs
- Half.cs
- IQuantGrids.cs
- KvBlockTransfer.cs
- KvCacheStorage.cs
- ManagedQuantizedOps.cs
- MediaHelper.cs
- ModelBase.cs
- Modelfile.fp16
- ModelMultimodalInjector.cs
- MoeCpuOffloadConfig.cs
- MtpSpeculativeDecoder.cs
- NativeDequant.cs
- PdfPageImageExtractor.cs
- PdfTextExtractor.cs
- TensorSharp.Models.csproj
- GbnfParser.cs
- Grammar.cs
- GrammarConstraint.cs
- GrammarLibrary.cs
- GrammarMatcher.cs
- GrammarTokenVocabulary.cs
- JsonSchemaGrammarCompiler.cs
- RegexToGbnf.cs
- FileLogger.cs
- FileLoggerOptions.cs
- FileLoggerProvider.cs
- FileLoggerScopeProvider.cs
- FileLogQueue.cs
- LogEntry.cs
- LogEntryFormatter.cs
- LogEventIds.cs
- LoggingExtensions.cs
- LogScopes.cs
- BlockHashIndex.cs
- BlockPool.cs
- BlockTable.cs
- KvBlock.cs
- ManagedPagedAttention.cs
- PagedKvBatchOps.cs
- PagedKvStorage.cs
- IRedisKeyValueStore.cs
- RedisConnection.cs
- BatchExecutor.cs
- ContinuousBatchScheduler.cs
- ExecutionCapabilities.cs
- ExecutionOptions.cs
- ExecutionPlan.cs
- ExecutionPlanner.cs
- InferenceEngine.cs
- InferenceRequestHandle.cs
- MtpSpeculativeExecution.cs
- SchedulerConfig.cs
- SchedulerOutput.cs
- SequenceState.cs
- SequenceStatus.cs
- BpeTokenizer.cs
- ChatTemplate.cs
- ConfigFileArgs.cs
- ConsoleBanner.cs
- Contracts.cs
- GgufReader.cs
- IKvBlockCodec.cs
- Jinja2Template.cs
- KvBlockHash.cs
- KVCache.cs
- KVCachePromptRenderer.cs
- ModelDownloader.cs
- ModelPrimitives.cs
- OutputParser.cs
- PagedKvBlockStore.cs
- PagedKvCacheConfig.cs
- PagedKvCacheManager.cs
- PrefillChunking.cs
- PromptRenderer.cs
- RecommendedSampling.cs
- RedisKvBlockTier.cs
- SafetensorsReader.cs
- SamplingConfig.cs
- SentencePieceTokenizer.cs
- SsdKvBlockTier.cs
- StructuredOutputs.cs
- TensorSharp.Runtime.csproj
- TokenSampler.cs
- TurboQuantKvCodec.cs
- Dockerfile_CPU.txt
- Dockerfile_GPU.txt
- README.md
- HealthEndpoints.cs
- OllamaEndpoints.cs
- OpenAIEndpoints.cs
- SessionEndpoints.cs
- UploadEndpoints.cs
- WebUiEndpoints.cs
- BackendSelector.cs
- HostedModelGuard.cs
- LoggingSetup.cs
- MtpStartupValidation.cs
- SamplingDefaults.cs
- ServerHostingOptions.cs
- ServerOptionsBuilder.cs
- ServerUsage.cs
- StartupBanner.cs
- StartupModelLoader.cs
- UploadCleanupService.cs
- UploadContentPolicy.cs
- UploadPathGuard.cs
- UploadStoragePolicy.cs
- WebRootSetup.cs
- ApiExceptionMiddleware.cs
- PromptOverflowMiddleware.cs
- RequestLoggingMiddleware.cs
- FinishReasonMapper.cs
- OllamaAdapter.cs
- OpenAIChatAdapter.cs
- OpenAIResponsesAdapter.cs
- WebUiAdapter.cs
- ChatMessageParser.cs
- SamplingConfigParser.cs
- ToolFunctionParser.cs
- UploadFileReference.cs
- WanVideoParamsParser.cs
- InMemoryResponsesStore.cs
- IResponsesStore.cs
- RedisResponsesStore.cs
- StoredResponse.cs
- JsonOptions.cs
- OllamaResponseFactory.cs
- OpenAIResponseFactory.cs
- OpenAIResponsesFactory.cs
- WebUiSseEvents.cs
- NdJsonWriter.cs
- SseWriter.cs
- StreamingJsonObjectFilter.cs
- README.md
- README_zh-cn.md
- test_multiturn.py
- test_multiturn.sh
- assistant_logo.png
- banner_1.png
- index.html
- .DS_Store
- API_EXAMPLES.md
- API_EXAMPLES_zh-cn.md
- BackendCatalog.cs
- ChatGenerationPipeline.cs
- ChatHistoryPreparer.cs
- ChatSession.cs
- DiffusionBatchScheduler.cs
- GlobalUsings.cs
- InferenceEngineHost.cs
- InferenceQueue.cs
- InferenceTelemetry.cs
- ModelLifecycleService.cs
- ModelService.cs
- OpenAIResponseFormatParser.cs
- Program.cs
- PromptContextOverflowException.cs
- SessionManager.cs
- TensorSharp.Server.csproj
- test_requests.jsonl
- TextUploadHelper.cs
- WebUiChatPolicy.cs
- .gitkeep
- README.md
- README_zh-cn.md
- MatrixConfig.cs
- MatrixConfigLoader.cs
- ModelDiscovery.cs
- matrix-config.json
- .gitkeep
- three_turn.jsonl
- audio_question.txt
- image_question.txt
- long_text.txt
- short_text.txt
- thinking_question.txt
- tools_question.txt
- upload_text.txt
- video_question.txt
- weather_tools.json
- Backend.cs
- EnvVarMatrix.cs
- FeatureCatalog.cs
- MatrixExpander.cs
- ModelSpec.cs
- TestCase.cs
- Baseline.cs
- BaselineStore.cs
- MarkdownReportGenerator.cs
- RegressionAnalyzer.cs
- ResultStore.cs
- TestResult.cs
- CliRunner.cs
- OutputParser.cs
- Program.cs
- README.md
- README_zh-cn.md
- TensorSharp.TestMatrix.csproj
- app.js
- from-tensors-to-tokens-cover.jpg
- logo.svg
- search-index-zh.js
- search-index.js
- style.css
- .nojekyll
- advanced.html
- advanced_zh-cn.html
- api-reference.html
- api-reference_zh-cn.html
- backends.html
- backends_zh-cn.html
- benchmarks.html
- benchmarks_zh-cn.html
- book.html
- book_zh-cn.html
- cli.html
- cli_zh-cn.html
- CNAME
- code-api.html
- code-api_zh-cn.html
- distributed.html
- distributed_zh-cn.html
- features.html
- features_zh-cn.html
- getting-started.html
- getting-started_zh-cn.html
- glossary.html
- glossary_zh-cn.html
- http-api.html
- http-api_zh-cn.html
- index.html
- index_zh-cn.html
- models-downloads.html
- models-downloads_zh-cn.html
- models-image.html
- models-image_zh-cn.html
- models-text.html
- models-text_zh-cn.html
- models-video.html
- models-video_zh-cn.html
- models.html
- models_zh-cn.html
- overview.html
- overview_zh-cn.html
- server.html
- server_zh-cn.html
- .gitattributes
- .gitignore
- DEVELOPMENT.md
- DEVELOPMENT_zh-cn.md
- Directory.Build.props
- Directory.Solution.props
- docs_models_README_zh-cn.md
- docs_models_wan_zh-cn.md
- FEATURES.md
- FEATURES_zh-cn.md
- LICENSE
- MODEL_DOWNLOADS.md
- MODEL_DOWNLOADS_zh-cn.md
- README.md
- README_zh-cn.md
- TENSOR_PARALLELISM_PLAN.md
- TensorSharp.slnx
- USAGE.md
- USAGE_zh-cn.md
# Installation Guide
git clone https://github.com/zhongkaifu/TensorSharp
Downloads the entire project code from GitHub to your computer.
cd TensorSharp
Moves into the project folder you just downloaded.
2. CMake
Medium Recommendedcd TensorSharp.Backends.MLX/Native
This project's files live in a subfolder, so move into it first.
mkdir build && cd build
Creates a folder to hold the build output and moves into it.
cmake ..
Analyzes the source code and generates build configuration files (must be run inside the build folder).
make
Compiles the code based on the generated build configuration to produce an executable.
3. .NET
Mediumdotnet --list-sdks
Type this command into your terminal and run it.
dotnet run --project TensorSharp.Cli -c Release -p:TensorSharpSkipMlxNative=true -- --model models\gemma-4-E4B-it-Q8_0.gguf --input prompt.txt --max-tokens 128 --backend ggml_cuda
Builds the project and runs it immediately.
add-apt-repository ppa:dotnet/backports
Type this command into your terminal and run it.
apt update && apt install dotnet-sdk-10.0
Type this command into your terminal and run it.
dotnet build -c Release
Type this command into your terminal and run it.
Pulled directly from this repo's README.
