SparkLLM

SparkLLM

Nov 7

A 4B model writes Python. Blender builds a bowl. What happens next?

#ai #opensource #python #machinelearning

5 min read

SparkLLM

Nov 6

A browser tab, a local AI: Spark-X2.6-1.7B on WebGPU

#webgpu #ai #opensource #javascript

3 min read

SparkLLM

Nov 4

Looks finished. Does it work? Six Spark X2.6 4B demos put to the test

#ai #javascript #opensource #webdev

1 min read

SparkLLM

Nov 4

An AI flagged the hiring clause. Its creators published the failures too.

#ai #opensource #a11y #community

2 min read

SparkLLM

Nov 3

Spark on your phone: 9 mobile app cases with sources and setup limits

#ai #opensource #android #ios

3 min read

SparkLLM

Nov 2

Spark-X2.6-4B in two live web assistants: share what you are building

#ai #opensource #llm #webdev

3 min read

SparkLLM

Nov 1

769 tok/s on a Single RTX 4090: Spark-X2.6-1.7B DSpark

#ai #opensource #machinelearning #llm

2 min read

SparkLLM

Oct 31

Spark-X2.6-4B DSpark Is Here: More Than 2x Average BF16 Inference Speed

#ai #opensource #machinelearning #llm

2 min read

SparkLLM

Oct 30

🎃 3.5M+ Downloads: Create Your Own Magic with Spark Open Models 🪄

#ai #opensource #machinelearning #halloween

2 min read

SparkLLM

Oct 30

4B, 94.6/100: Spark-X2.6 shines in a community NanoBench agent run

#ai #opensource #machinelearning #llm

2 min read

SparkLLM

Oct 27

Spark-X2.6-1.7B Is Here: Stronger Coding and Agents in a Compact Model

#ai #opensource #machinelearning #llm

3 min read

SparkLLM

Oct 26

Thank You for the Spark-X2.6-4B Community Feedback — Let’s Build for Good

#ai #opensource #machinelearning #hackathon

2 min read

SparkLLM

Oct 25

Spark-X2.6-4B reaches #4 on Hugging Face Trending: thank you, and build for good

#ai #opensource #machinelearning #hackathon

3 min read

SparkLLM

Oct 24

Spark X2.6 for Local Chinese-to-English Translation, With a Recorded Math Check

#ai #opensource #machinelearning #python

3 min read

SparkLLM

Oct 23

Turn barriers into access: HER Hack-Astron #7 is open

#ai #opensource #hackathon #machinelearning

2 min read

SparkLLM

Oct 20

Spark-X2.6-4B: stronger on-device agents, coding and controllable thinking

#ai #opensource #llm #machinelearning

2 min read

SparkLLM

Oct 16

Two roots, one valid size: a local Spark-X2.5-1.7B and SymPy demonstration

#ai #python #opensource #math

4 min read

SparkLLM

Oct 14

Checking an empty-input defect with Spark-X2.5 and OpenCodeReview locally

#ai #codereview #python #opensource

3 min read

SparkLLM

Oct 14

Spark X2.5 Passes 2 Million Downloads, and Spark-X2.6 Is Coming Soon

#ai #opensource #machinelearning #llm

3 min read

SparkLLM

Oct 12

Model calls first, model-off cache replay second: Spark-X2.5 tests SkillHub with e2e

#ai #testing #opensource #webdev

4 min read

SparkLLM

Sep 29

Spark X2.5 Passes 1 Million Downloads Across Official Channels and Community Builds

#ai #opensource #machinelearning #llm

2 min read

SparkLLM

Sep 11

Spark X2.5 Passes 100K Downloads on Hugging Face in 10 Days

#ai #opensource #machinelearning #llm

1 min read

SparkLLM

Sep 11

HER Hack-Astron #6 FAQ and a Spark-X2.5 case

#ai #opensource #machinelearning #testing

3 min read

SparkLLM

Sep 11

Spark-X2.5: Case 20 and Challenge FAQ

#ai #opensource #machinelearning #testing

2 min read

SparkLLM

Sep 9

Spark-X2.5 on Unsloth and LM Studio: a real bilingual local demo

#ai #opensource #machinelearning #tutorial

2 min read

SparkLLM

Sep 9

Right math, rejected answer? A Spark-X2.5 case worth reproducing

#ai #opensource #machinelearning #testing

2 min read

SparkLLM

Sep 8

New numbers, same math? A 12-problem Spark-X2.5 case and an invitation to test

#ai #opensource #machinelearning #testing

3 min read

SparkLLM

Sep 8

Spark’s 2180 vs GSM8K’s 2280: check the answer key, then join the challenge

#ai #opensource #machinelearning #testing

1 min read

SparkLLM

Sep 8

Right math, wrong final answer: a Spark-X2.5 community case and an invitation to test

#ai #opensource #machinelearning #testing

3 min read

SparkLLM

Sep 7

Spark-X2.5 hits #1 on HF Trending — join HER Hack-Astron #6: the Math Reasoning Challenge

#opensource #llm #hackathon #math

2 min read

SparkLLM

Sep 7

Spark-X2.5 runs locally: official llama.cpp support (Windows CPU demo)

#opensource #llm #llamacpp #localai

1 min read

SparkLLM

Sep 5

Can a 1.7B model really run end to end? — HER Hack-Astron #5

#opensource #llm #hackathon #edgeai

1 min read

SparkLLM

Sep 3

Spark-X2.5-4B: a 4B model matching 2–3 larger models on agents, code & math

#opensource #llm #benchmarks #ai

1 min read

SparkLLM

Sep 1

HER Hack-Astron #5: Build a Real Spark X2.5 Edge-Agent Use Case

#challenge #opensource #aiagents #edgeai

2 min read

SparkLLM

Sep 1

Spark X2.5-4B & 1.7B: the only on-device models with native 1M-token context — now open source

#opensource #llm #edgeai #ondevice

2 min read

SparkLLM

Aug 25

Domux: a compact open model for smart-home command understanding at the edge

#opensource #llm #edgeai #smarthome

1 min read

DEV Community

Badges

Writing Debut

A 4B model writes Python. Blender builds a bowl. What happens next?

A browser tab, a local AI: Spark-X2.6-1.7B on WebGPU

Looks finished. Does it work? Six Spark X2.6 4B demos put to the test

An AI flagged the hiring clause. Its creators published the failures too.

Spark on your phone: 9 mobile app cases with sources and setup limits

Spark-X2.6-4B in two live web assistants: share what you are building

769 tok/s on a Single RTX 4090: Spark-X2.6-1.7B DSpark

Spark-X2.6-4B DSpark Is Here: More Than 2x Average BF16 Inference Speed

🎃 3.5M+ Downloads: Create Your Own Magic with Spark Open Models 🪄

4B, 94.6/100: Spark-X2.6 shines in a community NanoBench agent run

Spark-X2.6-1.7B Is Here: Stronger Coding and Agents in a Compact Model

Thank You for the Spark-X2.6-4B Community Feedback — Let’s Build for Good

Spark-X2.6-4B reaches #4 on Hugging Face Trending: thank you, and build for good

Spark X2.6 for Local Chinese-to-English Translation, With a Recorded Math Check

Turn barriers into access: HER Hack-Astron #7 is open

Spark-X2.6-4B: stronger on-device agents, coding and controllable thinking

Two roots, one valid size: a local Spark-X2.5-1.7B and SymPy demonstration

Checking an empty-input defect with Spark-X2.5 and OpenCodeReview locally

Spark X2.5 Passes 2 Million Downloads, and Spark-X2.6 Is Coming Soon

Model calls first, model-off cache replay second: Spark-X2.5 tests SkillHub with e2e

Spark X2.5 Passes 1 Million Downloads Across Official Channels and Community Builds

Spark X2.5 Passes 100K Downloads on Hugging Face in 10 Days

HER Hack-Astron #6 FAQ and a Spark-X2.5 case

Spark-X2.5: Case 20 and Challenge FAQ

Spark-X2.5 on Unsloth and LM Studio: a real bilingual local demo

Right math, rejected answer? A Spark-X2.5 case worth reproducing

New numbers, same math? A 12-problem Spark-X2.5 case and an invitation to test

Spark’s 2180 vs GSM8K’s 2280: check the answer key, then join the challenge

Right math, wrong final answer: a Spark-X2.5 community case and an invitation to test

Spark-X2.5 hits #1 on HF Trending — join HER Hack-Astron #6: the Math Reasoning Challenge

Spark-X2.5 runs locally: official llama.cpp support (Windows CPU demo)

Can a 1.7B model really run end to end? — HER Hack-Astron #5

Spark-X2.5-4B: a 4B model matching 2–3 larger models on agents, code & math

HER Hack-Astron #5: Build a Real Spark X2.5 Edge-Agent Use Case

Spark X2.5-4B & 1.7B: the only on-device models with native 1M-token context — now open source

Domux: a compact open model for smart-home command understanding at the edge