The story so far
Weights, every dated article, and repo commits on one rail.
22 repo commits since the weights landed
Teal is the weights (2026-09-22), grey ticks are repo commits.
Lumma-fev-0.6b
FrontiersMind Apache 2.0 out 33 days ago
649M parameters, text in, labels out.
Verified 2026-10-25 Rank #0 in Decision models · score 0 method v1 Sources 3 Downloads 30d 9.9k · #507 of 619 tracked
Do people use it?9.9kdownloads, 30 days#507 of 619 we track · below the median+0 likes, downloads −8% this week68 likesHugging Face2026-10-24
Where it's used
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
No API provider serves it, so its downloads are people running it themselves: 4 Spaces.
- As a service0no API provider listed
- On own hardware0quantized builds
- Built on00 fine-tunes · 0 adapters · 0 merges
- In apps4Hugging Face Spaces
Will it fit?
Weights only, an estimate: leave room for the KV cache. Method
Fits a 16 GB card at INT4
About 0.4 GB at INT4, 0.8 GB at FP8, 1.6 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Reference
Everything else, folded: the full tables, every source.
Benchmarks
6 rows
Benchmarks 6 rows
measured third-party leaderboard rowauthor-reported number from the model card or eval files (6)No confidence intervals unless the source gave them; settings differ between rows, so compare within a benchmark, not across. Date = the benchmark's own date when it gives one, else when we read it.
| Benchmark | Value | Setting / notes | Reported by | Date |
|---|---|---|---|---|
| author-reported AG News | 0.84 | — | model author | 2026-10-07 |
| author-reported Average | 0.82 | — | model author | 2026-10-07 |
| author-reported Banking77 · Classify and make typed decisions | 0.9 | — | model author | 2026-10-07 |
| author-reported DAIR Emotion | 0.89 | — | model author | 2026-10-07 |
| author-reported P50 latency | 45.82 ms other | — | model author | 2026-10-07 |
| author-reported Typed-decisions | 0.64 | — | model author | 2026-10-07 |
What the card says
2 cited sentences
Summary generated, cited
Lumma-fev-0.6b is a 649M parameter decision model fine-tuned from the FrontiersMind/Lumma-0.6B-Base model.[1][2] The model performs text classification by returning a probability distribution for typed questions in a single forward pass.[3][4]
- parameters human
- “fine-tuned from FrontiersMind's Lumma-0.6B-Base”
- pipeline tag
- “returns a probability distribution for each question in a…”
Generated 2026-10-07 15:18 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
Use-case evidence
3 use cases
What it is good at 3 use cases
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Facts
18 fields
Listed in Decision models
Facts
- Lumma-fev-0.6b has 649M parameters. (safetensors metadata, observed 2026-10-24)
- License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed 2026-10-24)
- Created 2026-09-22, last modified 2026-10-02. (HF API, observed 2026-10-24)
| Pipeline tag | text-classification |
|---|---|
| Library | transformers |
| Architectures | FevForDecision |
| Model type | fev |
| Dtype | bfloat16 |
| Params by dtype | BF16: 648,544,684 · F32: 737,792 |
| Size band | S |
| MoE | no |
| Languages | en |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
Servability
vLLM · SGLang
Servability
vLLM: not in registry · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: not in registry · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Hardware fit
3 estimates
Hardware fit ESTIMATE
Weights-only memory for Lumma-fev-0.6b: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 1.6 GB · Edge (<4 GB) |
|---|---|
| FP8 (1 byte/param) | 0.8 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 0.4 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
Decision facts
catalog
Decision-model catalog curated
| How it was made | LoRA on own from-scratch pretrained base |
|---|---|
| Lineage bucket | O - original architecture or own base |
| Lineage evidence | model card: fine-tuned from Lumma-0.6B-Base "which we pre-trained from scratch" |
| Base model | FrontiersMind/Lumma-0.6B-Base |
| DI 0.2.1 score | 2.97 (independent (DI 0.2.1 board)) |
| ECE (board) | 0.2357 |
| Median latency | 21.3 ms (DI board, 1x RTX PRO 6000, single process) |
| Hosted price (input) | self-hosted/M |
| Contamination notes | 3,077 BANKING77 test items found in FrontiersMind's own decision datasets count as wrong |
| Source | https://huggingface.co/FrontiersMind/Lumma-fev-0.6b |
All variants
lineage
Lineage
Base model: FrontiersMind/Lumma-0.6B-Base
Repository changelog
22 commits
Repository changelog 22 commits
| Date | Commit | SHA |
|---|---|---|
| 2026-10-02 | Update README.md | 1b8506e |
| 2026-09-29 | Updated Readme.md (#4) | 59272e2 |
| 2026-09-28 | Updated benchmarks | d7a2390 |
| 2026-09-26 | Update README.md | 9d52716 |
| 2026-09-25 | Update README.md | 9f12b0c |
| 2026-09-25 | Update README.md | 6b0b0db |
| 2026-09-25 | Update README.md | 298ddf9 |
| 2026-09-25 | Update README.md (#3) | c735d16 |
| 2026-09-25 | Upload lumma-fev-0.6b | b53fae3 |
| 2026-09-23 | Update README.md (#1) | 23dc7bd |
| 2026-09-22 | Update README.md | 207ba26 |
| 2026-09-22 | Update README.md | f42f772 |
| 2026-09-22 | Update README.md | 8f9d79e |
| 2026-09-22 | Update README.md | 1c969b0 |
| 2026-09-22 | Update README.md | ecf68a5 |
| 2026-09-22 | Update README.md | 1c111a3 |
| 2026-09-22 | Update README.md | fd0f58b |
| 2026-09-22 | Update README.md | 308c313 |
| 2026-09-22 | Upload folder using huggingface_hub | 8cb555e |
| 2026-09-22 | Update README.md | 3da1e8e |
| 2026-09-22 | Upload folder using huggingface_hub | e747425 |
| 2026-09-22 | initial commit | f6fe4ae |
Similar models
3 models
Similar models
- mojev · 854M same size band (S), same task (text-classification), permissive license
- GLiNER2.5-Decide · 486M same size band (S), same task (text-classification), permissive license
- lavoir · 422M same size band (S), same task (text-classification)
Sources
3 sources
Sources
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:44 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:18 UTC · Last verified 2026-10-25
Follow Lumma-fev-0.6b
Watch this model
- /feeds/models/FrontiersMind--Lumma-fev-0.6b.xml(RSS 2.0) · every change to Lumma-fev-0.6b
- /feeds/families/FrontiersMind--Lumma.xml(RSS 2.0) · every change across the Lumma family
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.