
Senior Principal Architect at @VMwareTanzu / @Broadcom | Ex @Pivotal | Bichon Frise 🍋🐩
Tokyo, Japan
Joined April 2007
- Tweets36.6K
- Following201
- Followers3.7K
- Likes6.1K
Toshiaki Maki retweeted
GLM 5.3 Flash on 2x DGX Sparks v1.10 is out ⚡️
Lots of PRs merged and bugs squashed!
Here are some highlights:
- Works with Claude Code & Anthropic SDKs directly
- Much faster at very long chats (500k+ tokens)
- Opt-in disk cache: long chats resume in 0.2 s instead of 11 s, even after a restart
- Sampling is now as fast as greedy decoding
Full changelog here:
mia-ai.net/models/GLM-5.3-Fl…
TensorFold 0.6.3 is live with the Anthropic API, Control Room, CUDA video and a bunch of issues/PR’s merged.
Nemotron code edits on M5 Max now decode ~1.4x faster with one stream.
Thank you @MiaAI_lab @filicroval @bhcc2025 and every contributor and tester.
@volatilemarkts @plotarmordev
Toshiaki Maki retweeted
Also applies for GLM 5.3 Flash
Basically Anthropic confirmed my own conclusions that GLM is the most capable < 1T param model you can run locally.
Toshiaki Maki retweeted
Great article by @exolabs, @0xSero, and @alexocheema
Recommend read for anyone who plans getting into local AI, and owns or plans to own one or more DGX Sparks.
DGX Sparks are great little devices that punch well above their weight.
Toshiaki Maki retweeted
Come check out my website.
It includes a list of my latest & greatest recipes.
The "Install Model" button gives you a ready-to-use prompt for Claude, Codex, or others. It adapts the recipe to your setup, starts the server and tests it.
mia-ai.net/models
🚀 MCDMA is working between my Mac Studio & DGX Sparks, and now it's time to bring it to everyone who loves oMLX.
I’ve opened a new PR based on the Mac/CUDA cluster path I worked on a while back. It’s under review, and I can’t wait to hear people's feedback.
Additionally, I will be working directly with @MiaAI_lab to cook up some recipes for running distributed inference.
github.com/ashhart/mcdma
github.com/jundot/omlx/pull/…
Toshiaki Maki retweeted
If you don’t want to wire up your own Jev client in Spring Boot, I created a simple starter for you.
Add the dependency, set your API key, and inject JevClient. Built for Spring Boot 4 with RestClient and typed questions and answers.
github.com/danvega/jev-sprin…
Toshiaki Maki retweeted
🚀 Datasource Micrometer v2.3.0 and v1.5.0 are out!
Highlights:
• Customize when DataSource beans are instrumented
• Cache JdbcConnectionInfo per DataSource
Release notes:
github.com/jdbc-observations…
#SpringBoot #Micrometer #Observability #JDBC
Toshiaki Maki retweeted
This is for you single DGX Spark owners 💫
You can now run Qwen3.8 Flash NVFP4 on one DGX Spark with great performance!
- Up to 1M context, 1,431,164 KV cache
- Full image & video support
- 37 decode tok/s on prose single stream
- Up to 86 tok/s on prose 4 concurrent streams
- 1500-2000 tok/s prefill on any size.
- Stress tested with 400k prefill - passed.
IMO this is the BEST model to run now on a single dgx spark. It's better than Qwen3.8-27B, and also better than the 1x dgx spark version of DeepSeek v4 Flash.
Get it here:
github.com/MiaAI-Lab/Qwen3.8…
Toshiaki Maki retweeted
SoutherのWASM出力、できるようになりました! Southerで書いたドメインモデルを、Reactから呼ぶexampleも置いてます。
github.com/souther-lang/sout…
Toshiaki Maki retweeted
interesting post from RabbitMQ looking at RabbitMQ vs Apache Kafka
(though we might be a bit biased lol)
rabbitmq.com/docs/compare/ka…
Toshiaki Maki retweeted
A BIG moment for all DGX Spark users ⚡️
You can now run DeepSeek v4 Flash 0731 without needing a second unit, with quality high enough for reliable code generation, high context, and great speed!
Optimized for single stream session:
- EXL3 quantization
- 384k context (conservative) / ~440k kv cache (!)
- 47 tok/s single stream (structured)
- 1024 tok/s prefill
- 370k token needle test passed (super stable!)
Thanks @0xSero for this excellent EXL3 quant! Tuned for SparkInfer + DSpark speculative decoding, this is about the same quality of a Q4_K_M / Q5 GGUF!
All this is possible thanks to native NVFP4 KV cache & fixing kernel bugs in the upstream prefill path to make it work at all.
Get it here:
github.com/MiaAI-Lab/DeepSee…
Toshiaki Maki retweeted
on DGX Spark:
llama serve \
-hf ggml-org/Qwen3.8-27B-GGUF:Q4_K_M \
-hfd ggml-org/Qwen3.8-27B-GGUF:Q4_0 \
--spec-default \
--spec-type draft-mtp \
--reasoning-preserve --agent
Toshiaki Maki retweeted
⚡ If you have or want a DGX Spark, this post might be the most important one you come across this month.
Serve the BEST model available (DeepSeek V4 Flash 07-31, 284B parameters) on ONE Spark @ 1,000 tok/s prefill, 59 tok/s multi-agent serving. One command to install. 🧵
Toshiaki Maki retweeted
Hardening your Spring Boot app just got simpler. Tanzu Spring now offers SLSA L3 compliant Java dependency builds. Get cryptographically verifiable provenance for your entire dependency tree. Details here: brcm.tech/4wWD5so
Toshiaki Maki retweeted
Maven Dependency Plugin 3.11.0 is out, and it adds something Java developers have wanted for a long time:
Add a dependency from the command line.
$ mvn dependency:add -Dgav=org.apache.commons:commons-lang3
No copy/paste XML required.
maven.apache.org/plugins/mav…
#Java #Maven
Toshiaki Maki retweeted
I love Spring Boot so much. I love the autoconfiguration that provides useful out-of-the-box integrations with tons of libraries in an idiomatic, concise way. i love the incredibly rich production-worthy features. One of my favorite kinds of features, the kind that really drive home why a framework is a force-multiplier? It's when Spring Framework's "portable service abstractions" intersect with "cross-cutting concerns." Some examples:
- the new `InetAddressFilter` in Spring Boot 4.1 lets you restrict outbound HTTP requests, to limit for example calls to untrusted web sites. A simple enough contract, but we had to hook it into basically anywhere you might want to make an HTTP request (RestTemplate, RestClient, Apache HTTP Client, JDK HTTP Client, etc., etc.) in the entire portfolio.
- the `SslBundle` mechanism is a way to define and centralize "trust materials" like keys. Simple enough. But we had to then hook that in to _everything_ that could use it, or that even remotely has a notion of SSL. (Spoiler alert: basically everything does) This means: embedded webservers, REST clients, Cassandra, Couchbase, ElasticSearch, MongoDB, Redis, RabbitMQ (and RabbitMQ Streams), RSocket, etc. all now have consistent ways to secure them.
- ServiceConnections - want to live that `git-clone-run` life? Don't want to fiddle with configuration? Choose Docker Compose or Testcontainers on start.spring.io and then choose interesting middleware like RabbitMQ, PostgreSQL, etc., and Spring Boot will automatically start up a Docker Image and then connect to it without you doing a thing.
Is dealing with one of these things in isolation such a big deal? No, not at all. But it's the death by a thousand cuts. And a great framework like Spring Boot, with the consistency it - and only it - can impose, is the salve that heals.



















