explainx.ai350K+ learn from us
TrendingAI News TodayPathwaysSkills
Pricing
explainx.ai

Upskill in AI — 16 free pathways, live workshops & bootcamps, and 50+ courses from practitioners. Plus the skills, tools, and MCP servers to practice on.

follow us

follow on google

Add explainx.ai as a preferred source

corporate training

support@explainx.ai

get started

Find your pathTake Free Evaluation

community

Join the community

learn

mind: share how you thinkpathways — start freeworkshopsbootcampscoursescompare Explainxcertificationsmock testsexplainx universitycorporate traininglearn skills & mcp

discover

skillsmcp serversexplainx mcpestacktoolsmdx readeragentsllmsdesignsdictionarypeopleagi trackerfelony benchranks

company

aboutvisionmissionteaminstructorsteach on explainxpartnershipscommunityhackathonscareers

content

daily AI newsstate of AI — live resultsblogreleasespromptsgeneratorsresource libraryfor LLMsexplainx.ai kids

solutions

all solutionsdeveloper upskillingmarketing upskillingproduct manager upskillingleadership upskilling

newsletter · weekly

Get AI news, tools, and insights in your inbox.

supportcontactprivacytermsdata rightshow we create contentsubmission guidelines

© 2026 AISOLO Technologies Pvt Ltd

explainx.ai

On this page

  • TL;DR specs
  • Maxwell Block Sparse Attention (MBSA)
  • Training story (high level)
  • Benchmark highlights readers care about
  • Deployment notes
  • Tool calling and templates
  • Name collision warning
  • Limitations
  • When to pick Maxwell vs Qwen3.6 35B-A3B
  • MoE routing and mHC details
  • Long-context evaluation methodology
  • Agent and search benchmarks
  • Quantization and Hugging Face siblings
  • Licensing and compliance
  • Failure modes when serving
  • Research lineage: Muon and distillation
  • Who should download this week
  • Who should wait
  • Hacker News traction
  • Versioning and reproducibility
  • Tokenizer size note
  • Related reading
  • Fine-tuning and enterprise guardrails
  • Multilingual table takeaways
  • Roadmap signals to watch on Hugging Face
  • Memory and tensor parallel planning
  • Stress test prompts before you adopt
  • Governance and naming in procurement
← Back to blog

explainx / blog

Maxwell-V1-Flash: 35B MoE Open Weights on Hugging Face

Open Models, Hugging Face, MoE, Inference, Mixture of Experts

Part of Open-Weight Models

maxwell-model/Maxwell-V1-Flash is a 35B MoE text model with 3B active params, MBSA sparse attention, and Apache 2.0 weights on Hugging Face.

Nov 17, 2026·8 min read·Yash Thakker
add explainx.ai
go deep
Maxwell-V1-Flash: 35B MoE Open Weights on Hugging Face

Maxwell-V1-Flash is a newly trending open MoE language model on Hugging Face — about 35 billion total parameters with three billion active per token — distinct from the unrelated doubleAI “Maxwell” code-optimization brand. We reviewed the maxwell-model/Maxwell-V1-Flash model card on November 17, 2026, when it showed on the order of 100+ community likes and hundreds of downloads.

If you searched “Maxwell” after reading doubleAI’s GPU savings claims, reset your mental model: this Maxwell is an open text MoE with Apache 2.0 licensing and vLLM deployment notes, not a closed compiler service.

TL;DR specs

table · 2 cols
SpecValue (model card)
ArchitectureMoE causal LM
Total / active params35B / 3B (excl. embeddings)
Context256K tokens configured
AttentionMBSA block-sparse + dense prefix
Experts256 routed, 12 active; 2 shared
Layers45 transformer layers
LicenseApache 2.0

Maxwell Block Sparse Attention (MBSA)

The card claims each KV group keeps 8 prefix + 8 local 128-token blocks and adds up to 16 remote blocks chosen by a learned indexer — capping main attention work while retaining a full-history KV cache. They report the MBSA kernel is 6.7× faster than dense attention beyond the dense prefix in their internal tests (vendor figure — not independently verified here).

Long-context eval tables show RULER scores at 256K still in the low 90s on their base model row — competitive with Qwen3.5-base in the same table, per their numbers.

Training story (high level)

Public text on the card describes:

  • ~26T tokens multilingual pretraining with Muon on many matrices
  • Block-sparse attention introduced at 128K mid-training after indexer warmup
  • SFT + RL branches (math, code, agents) merged via multi-teacher on-policy distillation

Treat parameter totals and token counts as self-reported unless you reproduce from checkpoints.

Benchmark highlights readers care about

Post-trained Maxwell-V1-Flash rows in the card claim:

  • AIME 2026 mean@8 around 95.0 versus mid-90s rivals
  • SWE-Bench Verified (Claude Code harness) about 65.4% pass@1
  • BrowseComp up to 82.2% with context management enabled (footnote on card)
  • Long-context averages slightly ahead of Qwen3.6-35B-A3B on their table

Always read footnotes: some search scores include context management asterisks.

Deployment notes

The card requires NVIDIA driver ≥ 580 (CUDA 13) or VLLM_ENABLE_CUDA_COMPATIBILITY=1 in Docker. Download steps:

bash
export MODEL_DIR=$PWD/Maxwell-V1-Flash
hf download maxwell-model/Maxwell-V1-Flash --local-dir "$MODEL_DIR"

Then run the documented vLLM OpenAI-compatible server docker one-liner from the card (adjust GPUs and shm).

Compare hosting economics to Baseten frontier inference guidance and Together agent eval stacks.

Tool calling and templates

The Hugging Face chat_template_jinja exposes <tool_call>, <think>, and multi-turn tool responses — aligned with agent harnesses like Qwen Code and Spillway guard patterns.

Name collision warning

table · 2 cols
NameWhat it is
Maxwell-V1-FlashOpen MoE LM on HF
doubleAI MaxwellCode execution optimizer (our prior post)
NVIDIA MaxwellGPU architecture generation (historical)

Documentation and bookmarks should use org slug maxwell-model to avoid mixing repos.

Limitations

  • VRAM: 35B MoE BF16 is not a laptop model.
  • Evals: Vendor tables, not explainx.ai reruns.
  • Org transparency: Maxwell Team is newer than tier-one labs — check commit history and quant releases before production bets.

When to pick Maxwell vs Qwen3.6 35B-A3B

Choose Maxwell if you need 256K sparse attention behavior and Apache weights on your vLLM cluster today. Choose Qwen if you need wider hosted API support and corporate support channels. Many teams will download both and run identical agent prompts — same playbook as K2 Horizon fleet posts.

MoE routing and mHC details

The spec table on the card lists 256 routed experts with 12 active per token, plus 2 shared experts per layer, using mHC (multi-head compressed) residual streams with Sinkhorn–Knopp iterations for routing stability. That is a more exotic stack than early Mixtral clones — inference engines must support custom code (custom_code tag on HF).

If your vLLM build lacks Maxwell kernels, expect fallback paths or build failures until the team publishes compatible versions.

Long-context evaluation methodology

Base-model RULER rows show graceful decay from 8K to 256K, with Maxwell-V1.0-Flash (Base) staying above 93 even at 256K in their table while Qwen3.5-base drifts lower. HELMET scores trail RULER — normal for composite long tasks — but still trend upward versus the Qwen row at 128K.

Post-trained models switch evaluation lengths: 128K for math and HLE, 80K for science, 64K for LiveCodeBench v6, per card footnotes. Reproducers must match generation caps or comparisons lie.

Agent and search benchmarks

The card includes Terminal-Bench, SWE-Bench Verified with Claude Code 2.1.212, Tau3-Bench, MCP-Mark, and a suite of search QA sets (BrowseComp, DeepSearchQA, WideSearch). Footnoted context management boosts some search passes — read asterisks before quoting 82% BrowseComp.

These harnesses align Maxwell marketing with agent product wars covered in Together eval posts and Google GTIG exploit agents — different problem, same buyer psychology.

Quantization and Hugging Face siblings

The Hugging Face model card flags quantized variants (hasQuantizations: true in page metadata when we fetched it), but quant artifacts may lag the BF16 drop. Watch the maxwell-model org for GGUF or FP8 variants before planning single-GPU home labs.

Licensing and compliance

Apache 2.0 enables commercial fine-tunes, but downstream benchmark claims in marketing still require reproduction. Legal teams care; ML engineers should too.

Failure modes when serving

table · 2 cols
IssueMitigation
Driver < 580Set VLLM_ENABLE_CUDA_COMPATIBILITY=1 per card
OOM on 24 GBUse published quants or tensor parallel
Tool template driftPin tokenizer revision on HF commit hash
Slow MoEVerify expert parallelism flags in vLLM

Research lineage: Muon and distillation

Pretraining notes cite Muon optimizer usage on hidden matrices — timely alongside Echelon MoE training research. Multi-teacher on-policy distillation after independent RL branches mirrors industry pipeline stories without naming Maxwell as a product company.

Who should download this week

  • Agent startups needing an Apache weights baseline at ~35B MoE scale.
  • Inference vendors testing sparse attention kernels against Qwen3.6.
  • Students comparing chat templates with tool XML to MiniMind or Qwen Code configs.

Who should wait

Teams needing hosted SLAs, safety classifiers, or EU AI Act paperwork from a major lab should stay on Alibaba Qwen4 API tiers until Maxwell publishes system cards and red-team results.

Hacker News traction

A linked HN item had minimal comments when we checked. Engagement is primarily Hugging Face likes and GitHub technical report links on the card, including the official GitHub repository and its technical report PDF.

Versioning and reproducibility

Pin lastModified dates from the HF card when filing internal model cards. Maxwell-V1-Flash updated in mid-November 2026; silent card edits can change chat templates without a semver tag. Mirror weights to private object storage if your compliance policy forbids runtime pulls from Hugging Face.

Tokenizer size note

The model card lists 200,704 vocabulary entries, much larger than MiniMind’s 6.4k didactic tokenizer. Embedding tables consume meaningful disk even when only 3B parameters activate per forward pass. Budget extra NVMe for shard downloads before overnight pulls.

If your org already standardized on Qwen3.6 tool schemas, budget an afternoon to map Maxwell XML tool blocks into your harness before flipping production traffic. Small template mismatches look like mysterious refusals in logs even when weights load fine.

Daily briefFrom Yash · 350K+ students taught

Catch up on AI

One AI story explained in plain English, plus one thing to try. Every day, free. Unsubscribe anytime.

Related reading

  • doubleAI Maxwell claimed vs verified
  • Qwen Code harness guide
  • Tencent HY4 open weights plans
  • Together AI coding agent evals
  • Baseten efficient inference
  • Hugging Face OpenRouter routing
  • Sarvam open-weight India stack
  • Echelon MoE training optimizer

Fine-tuning and enterprise guardrails

Apache weights invite LoRA on domain corpora, but Maxwell’s tool template is complex — fine-tunes must preserve XML tool syntax or break agent harnesses. Enterprise buyers should still run internal red-team suites before exposing Maxwell to customer PII, matching discipline we describe for OpenAI CSAM classifier rollouts and Spillway injection guards.

Multilingual table takeaways

The card’s multilingual average lands near 89.6 post-trained, with strong MMMLU and MGSM rows. Teams serving SEA languages should compare against Sarvam India models and Tencent HY4 on their own eval harness, not vendor tables alone.

Roadmap signals to watch on Hugging Face

Watch for: technical report PDF link on the org page, quantized repos, vLLM version pins, and Model index entries tying Base vs Flash vs Instruct variants. Missing pieces mean paper + weights preview, not finished product.

Memory and tensor parallel planning

At ~736 GB total file size on the model card for BF16 shards, plan tensor parallel width before promising latency SLAs. Multi-GPU nodes should reserve headroom for KV cache growth at 256K context — agent sessions blow past 32K quickly when tool outputs recurse.

For cost modeling, compare $/1M tokens against hosted Qwen API tiers and against RightNOW kernel inference pricing if you only need narrow tasks.

Stress test prompts before you adopt

Run your production JSON tool schema, grep-heavy code edit, and long-context RAG prompts side by side against Qwen3.6-35B-A3B on the same GPU count. Maxwell’s win condition is better accuracy at equal cost, not mystique.

Governance and naming in procurement

Procurement spreadsheets should list maxwell-model/Maxwell-V1-Flash explicitly so legal does not confuse the download with doubleAI services or NVIDIA Maxwell GPUs sitting in the same rack row.

Primary sources: maxwell-model/Maxwell-V1-Flash model card, accessed November 17, 2026.

Spotted something out of date? Let us know.
Daily briefFrom Yash · 350K+ students taught

Catch up on AI

One AI story explained in plain English, plus one thing to try. Every day, free. Unsubscribe anytime.

  • ✓ Free
  • ✓ One email a day, no spam
  • ✓ Unsubscribe anytime
Yash Thakker

Written by

Yash Thakker

Yash is an AI expert with over 300K learners. Join his workshops →

View Yash Thakker in People in AI →

Related posts

Nov 16, 2026

Hugging Face Routes Hub Inference Through OpenRouter: One Token, Many Providers

Hugging Face and OpenRouter announced on November 16, 2026 that language-model inference on the Hub now routes through OpenRouter instead of a custom router. Builders keep one Hugging Face token while OpenRouter picks among 80+ providers by price, speed, and reliability.

Nov 17, 2026

RightNow AI YC Launch: Kernel-Optimized Open-Model Inference API

RightNow emerged from Y Combinator’s Fall 2026 batch with a Launch YC post on November 17 positioning the company as a GPU kernel lab serving open-weight models through drop-in OpenAI and Anthropic client libraries. Published prices start at $0.05 per million input tokens, with live cache-hit rates on the pricing page and zero-retention defaults in docs.

Nov 13, 2026

InferSSD: Running 800B-Class MoE Models From NVMe on Consumer GPUs

InferSSD is an open llama.cpp-based runtime that stores routed mixture-of-experts weights on NVMe while attention and KV cache stay in RAM or VRAM. WiCi Research published benchmark tables for DeepSeek-V4.1-Flash-1017 and Muse-Spark-Open on RTX 5090, MacBook Pro, and DGX Spark class machines.