←Search results

Search “NVIDIA”

By relevance · Total 89 results Clear search

Wednesday

  1. NVIDIA Developer · Generative AI

    NVIDIA Explains How to Fine-Tune Nemotron 3.5 Lightning with the Nemotron-Personas Dataset

    NVIDIA has released the Nemotron-Personas France and Belgium datasets and provided a complete post-training workflow for fine-tuning Nemotron 3.5 Lightning with them. The workflow includes defining behavior contracts, generating synthetic conversations with personas, fine-tuning with LoRA on NVIDIA NeMo, merging adapters, and then comparing and evaluating item by item against the base model on a frozen test set. The France dataset was built in collaboration with Pleias, and the Belgium dataset involved Pleias and KU Leuven; the Nemotron repository provides reusable steps and agent skills for data generation, training, and evaluation.

    Flowchart showing eight stages: define target behavior and a holdout set, sample personas, generate conversations, audit and split data, fine-tune with LoRA, merge the adapter, serve the adapted model, and evaluate it against the base model.

    NVIDIA has released the Nemotron-Personas France and Belgium datasets and provided a complete post-training workflow for fine-tuning Nemotron 3.5 Lightning with them. The workflow includes defining behavior contracts, generating synthetic conversations with personas, fine-tuning with LoRA on NVIDIA NeMo, merging adapters, and then comparing and evaluating item by item against the base model on a frozen test set. The France dataset was built in collaboration with Pleias, and the Belgium dataset i

    Matched fieldsTitle、Summary、Body

    #lab#Tutorials / practice#English text is being prepared.#Open-source ecosystem

Monday

  1. n8n Blog

    How to run n8n AI workflows on NVIDIA DGX Spark without calling cloud APIs

    n8n Blog published a tutorial on how to deploy a complete model stack locally on a single NVIDIA DGX Spark, enabling n8n workflows to perform transcription, OCR, document Q&A, and image generation without calling cloud APIs. The entire stack exposes an OpenAI-compatible endpoint through LiteLLM; the main model is the NVFP4 version of Qwen3.6-35B-A3B, with a single-user generation speed of 120 tokens/s and an aggregate of 359 tokens/s across eight users. The author's testing found that image generation reduces chat speed to about one-third, and recommends that small teams queue image tasks or schedule them outside working hours.

    n8n Blog published a tutorial on how to deploy a complete model stack locally on a single NVIDIA DGX Spark, enabling n8n workflows to perform transcription, OCR, document Q&A, and image generation without calling cloud APIs. The entire stack exposes an OpenAI-compatible endpoint through LiteLLM; the main model is the NVFP4 version of Qwen3.6-35B-A3B, with a single-user generation speed of 120 tokens/s and an aggregate of 359 tokens/s across eight users. The author's testing found that image gene

    Matched fieldsTitle、Summary、Body

    #engineering#Tutorials / practice#English text is being prepared.#Open-source ecosystem

Friday

  1. heise online

    RTX Spark laptop tested: How does Nvidia's first local AI laptop chip perform?

    heise online's c't 3003 tested the Asus ProArt P14, which is equipped with Nvidia's first laptop chip, the N1X (brand name RTX Spark). The machine supports up to 128 GB of unified memory; the test unit had 32 GB of memory plus a 1 TB SSD and cost 3,300 euros. In llama-bench tests with Qwen 3.8-27B, generation speed was about 14.5 tokens/s, slightly faster than Apple M6 and Strix Halo but slower than M5 Pro; with the MoE model Qwen 3.6-35B-A3B, it reached nearly 61 tokens/s. Prompt processing speed was better than Strix Halo and the M5 Pro and M6, but fan noise under full load reached up to 2.7 sone, and the fans also spin up when idle. The author also tried installing Ubuntu; after manually adding the drivers, both models ran 14% and 9% faster under Linux than under Windows.

    heise online's c't 3003 tested the Asus ProArt P14, which is equipped with Nvidia's first laptop chip, the N1X (brand name RTX Spark). The machine supports up to 128 GB of unified memory; the test unit had 32 GB of memory plus a 1 TB SSD and cost 3,300 euros. In llama-bench tests with Qwen 3.8-27B, generation speed was about 14.5 tokens/s, slightly faster than Apple M6 and Strix Halo but slower than M5 Pro; with the MoE model Qwen 3.6-35B-A3B, it reached nearly 61 tokens/s. Prompt processing spe

    Matched fieldsTitle、Summary、Body

    #hot-signal#Evaluations / benchmarks#English text is being prepared.#English text is being prepared.
  2. Computerworld

    Surface Laptop Ultra review: NVIDIA RTX Spark's local AI performance

    The author tested Microsoft's Surface Laptop Ultra equipped with the NVIDIA RTX Spark N1X SoC. The top configuration of this machine features a 20-core Grace Arm CPU, a 6144-core Blackwell GPU, and 128GB of unified memory, with a starting price of $2,599.99. In the UL Procyon AI text generation test, it scored 2703, about 33% higher than the IdeaPad Pro 5 with an Intel Core Ultra X9 388H; MLPerf Flux 2 Klein 4B image generation reached 26.97 images per minute, about 2.7 times that of the latter. The author also used a local Qwen3.8-27B to drive the Hermes agent, which completed extracting benchmark charts from Excel and generating a Word report in about five minutes.

    The author tested Microsoft's Surface Laptop Ultra equipped with the NVIDIA RTX Spark N1X SoC. The top configuration of this machine features a 20-core Grace Arm CPU, a 6144-core Blackwell GPU, and 128GB of unified memory, with a starting price of $2,599.99. In the UL Procyon AI text generation test, it scored 2703, about 33% higher than the IdeaPad Pro 5 with an Intel Core Ultra X9 388H; MLPerf Flux 2 Klein 4B image generation reached 26.97 images per minute, about 2.7 times that of the latter.

    Matched fieldsTitle、Summary、Body

    #hot-signal#Evaluations / benchmarks#English text is being prepared.#Agents

Thursday

Wednesday

Friday

Tuesday

  1. Chips and Cheese

    NVIDIA Olympus Core Microarchitecture Tested: Pushing the Server Single-Thread Ceiling with Per-Clock Performance

    Chips and Cheese conducted microarchitecture testing on NVIDIA's server core Olympus, a 10-wide out-of-order design clocked at 3.3 GHz that boosts per-clock performance by enlarging out-of-order structures and SMT. In SPEC CPU2026, its single-thread performance approaches that of Intel and AMD desktop cores and clearly leads Neoverse V2 and N2, but the test system used 64 KB pages, creating bias when compared with other 4 KB page results. The author notes that its spatial multithreading statically splits resources such as L1, L2, and scheduling queues in half; gains on integer workloads are comparable to Zen 5, while throughput declines on workloads limited by the floating-point backend.

    Chips and Cheese conducted microarchitecture testing on NVIDIA's server core Olympus, a 10-wide out-of-order design clocked at 3.3 GHz that boosts per-clock performance by enlarging out-of-order structures and SMT. In SPEC CPU2026, its single-thread performance approaches that of Intel and AMD desktop cores and clearly leads Neoverse V2 and N2, but the test system used 64 KB pages, creating bias when compared with other 4 KB page results. The author notes that its spatial multithreading statical

    Matched fieldsTitle、Summary、Body

    #hot-signal#English text is being prepared.#English text is being prepared.#Evaluations / benchmarks

Saturday

Thursday

Wednesday

Friday

Thursday

Thursday

Saturday

  1. Artificial Analysis

    Artificial Analysis measured Meta Muse Spark Open inference throughput

    Artificial Analysis published preliminary test results on AA-AgentPerf for Meta's new open-weight model Muse Spark Open, an 816B-parameter, 42B-active MoE that scores 39 on the Artificial Analysis Intelligence Index. Tests show NVIDIA B300 and GB300 each reach about 150M tokens/s per megawatt under throughput-first configurations, with GB300 at about 108M and B300 at about 92M at 100 tok/s output speed; AMD MI355X reaches 70M tokens/s per megawatt at 65 tok/s. 46 of the model's 62 layers use 2,048-token sliding window attention, reducing KV cache requirements at 1M context by about 75% versus full attention; Meta released an NVFP4 checkpoint and AMD released an MXFP4 checkpoint.

    esults on AA-AgentPerf for Meta's new open-weight model Muse Spark Open, an 816B-parameter, 42B-active MoE that scores 39 on the Artificial Analysis Intelligence Index. Tests show NVIDIA B300 and GB300 each reach about 150M tokens/s per megawatt under throughput-first configurations, with GB300 at about 108M and B300 at about 92M at 100 tok/s output speed; AMD MI355X reaches 70M tokens/s per megawatt at 65 tok/s. 46 of the model's 62 layers use 2,048-token sliding window attention, reducing KV c

    Matched fieldsSummary、Body

    #editorial#Evaluations / benchmarks#Reasoning#English text is being prepared.
  2. SemiAnalysis

    SemiAnalysis benchmarks Trainium3's PyTorch training and Neuron Explorer profiler

    The SemiAnalysis engineering team benchmarked the native PyTorch training stack TorchNeuron and TorchTitan on Trainium3, training Qwen3-32B on 16 Trainium2 chips; eager mode reached 11,099 tokens/s and 23.89% MFU, while compiled mode reached 14,156 tokens/s and 30.47% MFU. The author considers the Neuron Explorer profiler superior to Nvidia's and AMD's solutions, but first-time compilation takes too long: the first-step compilation for the gpt-oss training recipe took 27 minutes 17 seconds. FlexAttention tests passed 89 of 96 cases, with all 7 failures due to compilation timeouts rather than numerical errors.

    um3, training Qwen3-32B on 16 Trainium2 chips; eager mode reached 11,099 tokens/s and 23.89% MFU, while compiled mode reached 14,156 tokens/s and 30.47% MFU. The author considers the Neuron Explorer profiler superior to Nvidia's and AMD's solutions, but first-time compilation takes too long: the first-step compilation for the gpt-oss training recipe took 27 minutes 17 seconds. FlexAttention tests passed 89 of 96 cases, with all 7 failures due to compilation timeouts rather than numerical errors.

    Matched fieldsSummary、Body

    #analysis#Industry news#English text is being prepared.#English text is being prepared.

Wednesday