100x faster than generative LLMs • Runs on laptops & cloud CPUs • Global #1 on JevBench When you ask ChatGPT or Claude a question, it generates words one token at a time, like a person typing out an essay. That takes 2 to 5 seconds and burns expensive GPU compute. That is great for writing a story, but it is painfully slow and expensive for simple decisions: - "Did the AI make up this answer, or is it actually in the PDF?" - "Should this customer's message go to billing, shipping, or technical support?" - "Does the revenue bar chart support this financial claim?" - "Did the student get the math problem right according to the answer key?" Psychologist Daniel Kahneman described human thinking…
messier-one is an open-weight model for text classification from Agentmessier, released under Creative Commons Attribution-NonCommercial 4.0. It has 5.3B parameters and a 262,144-token context. At 16-bit it needs about 12.7 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 115 downloads a month.
A decision model: it reads a state and typed questions (choice with 2-255 options, score with 2-10 levels, noul) and returns a probability for every option from one forward pass. It does not generate text.
Parameters5.3B
Context262,144
Weights10.6 GB
Licensecc-by-nc-4.0
AccessOpen weights
Monthly Downloads115
Runs On
What it takes to serve messier-one (5.3B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 10.6 GB | 12.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 5.3 GB | 6.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 2.6 GB | 3.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 22, 2026.
messier-one on every accelerator the SAVRN Index prices, at every precision
Model Card
A decision model: it reads a state and typed questions (choice with 2-255 options, score with 2-10 levels, noul) and returns a probability for every option from one forward pass. It does not generate text. The wire format is TypeSafe-compatible (POST /v1/systemone); the serving code is at https://github.com/agentmessier-ai/messier-one. - One forward pass, no generated text. Every answer is a probability for each option, read at one position. On one RTX 3090 a decision takes about 80 ms (short documents) to 145 ms (long ones), server on loopback. - Up to 255 options in one question. Routing to one of many tools, picking a category from a long list, choosing a move among many: on our own…
Excerpt from the card by Agentmessier, licensed cc-by-nc-4.0.
Configuration
- Architecture
- Qwen3_5ForConditionalGeneration
- Context length (tokens)
- 262,144
- Layers
- 32
- Hidden size
- 2,560
- Feed-forward size
- 9,216
- Attention heads
- 16
- Key/value heads
- 4
- Head dimension
- 256
- Vocabulary size
- 248,320
- Model type
- qwen3_5
Identity and Version
- Repository
- agentmessier/messier-one
- Publisher
- Agentmessier
- Task
- Text classification
- Modality
- Text
- Library
- transformers
- Parameters
- 5.3B parameters
- Languages
- Not stated by the source
- Revision
- 9e6c71537e578c6e6311c5767dc1a5a31fa2e374
- First published
- 2026-10-06
- Last updated
- 2026-10-19
Files and Weights
17 files, 10.6 GB in total. The weights are 2 files totalling 10.6 GB in safetensors.
Weights2 files · 10.6 GB
Configuration6 files · 82.1 KB
Tokenizer4 files · 22.9 MB
Documentation3 files · 17.4 KB
Other1 file · 7.8 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors-00001-of-00002.safetensors | Weights | 6.6 GB | ad8891f9dbf7 |
| model.safetensors-00002-of-00002.safetensors | Weights | 4.0 GB | d7168057fce0 |
| config.json | Configuration | 2.7 KB | — |
| model.safetensors.index.json | Configuration | 76.3 KB | — |
| preprocessor_config.json | Configuration | 390 B | — |
| sys1_merge.json | Configuration | 2.3 KB | — |
| temps.json | Configuration | 61 B | — |
| video_preprocessor_config.json | Configuration | 385 B | — |
| LICENSE | Documentation | 575 B | — |
| LICENSE-Qwen-Apache-2.0 | Documentation | 11.5 KB | — |
| README.md | Documentation | 5.2 KB | — |
| chat_template.jinja | Other | 7.8 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 3.4 MB | — |
| tokenizer.json | Tokenizer | 12.8 MB | 5f9e4d4901a9 |
| tokenizer_config.json | Tokenizer | 16.7 KB | — |
| vocab.json | Tokenizer | 6.7 MB | — |
License and Download
- License
- cc-by-nc-4.0
- Access
- Open weights, no gate
- Download size
- 10.6 GB
Download from Agentmessier
Released by Agentmessier through its official repository on Hugging Face. Read the license.
Built From
- Derived from Qwen/Qwen3.5-4B
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 10.6 GB |
| 16-bit | 10.6 GB |
| 8-bit | 5.3 GB |
| 4-bit | 2.6 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About messier-one
How much GPU memory does messier-one need?
About 12.7 GB at 16-bit and 3.2 GB at 4-bit: the weights (5.3B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run messier-one on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use messier-one commercially?
Not without separate permission. messier-one is released under Creative Commons Attribution-NonCommercial 4.0. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only. Commercial use needs separate permission from the rights holder.
What is messier-one's context length?
262,144 tokens, from the maximum position embeddings in its published configuration.
Similar Models
100x faster than generative LLMs • Runs on laptops & cloud CPUs • Global #1 on JevBench When you ask ChatGPT or Claude a question, it generates words one token at a time, like a person typing out an essay. That takes 2 to 5 seconds and burns expensive GPU compute. That is great for writing a story, but it is painfully slow and expensive for simple decisions: - "Did the AI make up this answer, or is it actually in the PDF?" - "Should this customer's message go to billing, shipping, or technical support?" - "Does the revenue bar chart support this financial claim?" - "Did the student get the math problem right according to the answer key?" Psychologist Daniel Kahneman described human thinking…
Experimental open-weights judgment model by Kitani OpenJudgement is unfinished. We're releasing this checkpoint for people to experiment with, inspect, and build on. It still needs work on judgment quality, calibration, and inference efficiency. It is not as good as Jev overall in our internal task comparisons. It does show a meaningful improvement over untouched Qwen on our recorded validation comparison: 75.4% versus 64.2% annotation agreement. That is a result on a particular evaluation set, not a claim that we beat the base model on every task. There are questions it handles well and questions it confidently gets wrong. Please judge the preview by your own examples rather than assuming…
A calibrated typed-decision model: give it a state (text, ticket, policy, JSON) and a typed question — choice, boolean, or rubric score — and it returns a probability for every option in a single forward pass (~24 ms). No generation, no parsing, nothing to hallucinate. 12 of 13 subsets exceed Bespoke Nimble-9B — a model 2.2× its size — same prompt format, same scoring protocol. The primary suite: BoolQ, MultiNLI, PAWS, PubMedQA, SQuAD-2, VitaminC, Civil Comments, Aegis 2.0, MASSIVE (en/de), HelpSteer-2, SummEval (consistency / relevance). Every item human-labeled; byte-reproducible (manifest-locked ids + sha256); same protocol as the Bespoke Nimble evaluation. Wins: verification-style noul…
uratori-ja-4b is a Japanese decision model. Given a text state (for example a source document and a claim) and one or more typed questions, it returns a probability distribution over the allowed answers in a single forward pass, without generating text. Three question types are supported: noul (yes/no), choice (pick one of 2 to 8 labelled options) and score (an ordinal scale with 2 to 10 levels). It is trained for three jobs: checking whether a claim or an answer is supported by supplied evidence (grounding), comparing two documents (contradictions, meaning-changing edits, added claims) and judging retrieval results and answers in a RAG pipeline (relevance, sufficiency, faithfulness). The…