Up to date
LiquidAI lfm1.0 out 24 days ago
2.17B parameters, several kinds of input and output.
Verified 2026-11-15 06:56 UTC Rank #434 in LLMs · score 77.6 method v1 Sources 3 Downloads 30d 21k · #449 of 618 tracked
Changed this week: 2 new quantized builds on the Hub (42 in all).2026-11-14 · 1 new quantized build on the Hub (40 in all).2026-11-13 · 1 new quantized build on the Hub (39 in all).2026-11-12 · 3 more
Picked by rule from 2 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The 1 pick above, plus these 1. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
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: 42 quantized builds.
Weights only, an estimate: leave room for the KV cache. Method
About 1.3 GB at INT4, 2.6 GB at FP8, 5.2 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2026-10-21), amber triangles are 2 dated articles (each links to it).
64 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
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.
Listed in LLMs
| Pipeline tag | any-to-any |
|---|---|
| Library | transformers |
| Architectures | Lfm3OmniForConditionalGeneration |
| Model type | lfm3_omni |
| Params by dtype | BF16: 2,166,928,384 |
| Size band | M |
| MoE | no |
| License | lfm1.0 (other) |
| Hub snapshot | 2026-11-15 |
Weights-only memory for LFM3-Omni-2B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 5.2 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 2.6 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 1.3 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
quantized: 42finetune: 21merge: 1
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| LiquidAI/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | LiquidAI | 174 | 447.1k |
| ggml-org/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | ggml-org | 17 | 24.5k |
| abliter8-ai/lfm3-2b-omni-abliter8-gguf | quantized | gguf | — | — | other | abliter8-ai | 18 | 11.5k |
| LiquidAI/d1-Omni-2B | finetune | — | 2.17B | M | other | LiquidAI | 109 | 4.8k |
| AtomicChat/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | AtomicChat | 3 | 7.1k |
| LiquidAI/LFM3-Omni-2B-MLX-4bit | quantized | MLX | 2.17B | M | other | LiquidAI | 11 | 4.7k |
| LiquidAI/LFM3-Omni-2B-MLX-8bit | quantized | MLX | 2.17B | M | other | LiquidAI | 10 | 3.9k |
| LiquidAI/LFM3-Omni-2B-MLX-bf16 | quantized | MLX | 2.17B | M | other | LiquidAI | 10 | 3.4k |
| PollardWeights/LFM3-Omni-2B-Pollard | quantized | GGUF | — | — | other | PollardWeights | 2 | 3.3k |
| IndrasMirror/LFM3-Omni-2B-XQ-Uncensored | quantized | GGUF | — | — | other | IndrasMirror | 2 | 2.6k |
| LiquidAI/LFM3-Omni-2B-ONNX | quantized | — | — | — | other | LiquidAI | 16 | 1.8k |
| erenyeager-1/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | erenyeager-1 | 0 | 1.9k |
| JonusNattapong/LFM3-Omni-2B-abliterated | finetune | — | 2.17B | M | other | JonusNattapong | 1 | 1.8k |
| prithivMLmods/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | prithivMLmods | 2 | 1.6k |
| vladimir94/lfm3-omni-2b-ane-coreml | quantized | — | — | — | other | vladimir94 | 0 | 1.6k |
| mizahnyx/lfm3-omni-2b-abliterated-heretic-v2 | quantized | GGUF | 2.17B | M | other | mizahnyx | 1 | 1.4k |
| NANI-Nithin/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | — | NANI-Nithin | 0 | 1.4k |
| Soulfate24/LFM3-Omni-2B-Paretrix-GGUF | quantized | GGUF | — | — | other | Soulfate24 | 0 | 1.4k |
| OpenIntelligenceNet/LFM3-Omni-2B-Uncensored | finetune | — | 2.17B | M | — | OpenIntelligenceNet | 8 | 931 |
| mradermacher/LFM3-Omni-2B-i1-GGUF | quantized | GGUF | — | — | other | mradermacher | 1 | 1.3k |
| infosave/LFM3-Omni-2B-cmf | quantized | — | — | — | other | infosave | 1 | 969 |
| litert-community/LFM3-Omni-2B-LiteRT | quantized | — | — | — | other | litert-community | 11 | 426 |
| mizahnyx/lfm3-omni-2b-abliterated-heretic-v2-GGUF | quantized | GGUF | — | — | other | mizahnyx | 0 | 860 |
| ragraph-ai/LFM3-Omni-2B-RP-heretic | finetune | GGUF | 2.17B | M | other | ragraph-ai | 2 | 731 |
| ragraph-ai/LFM3-Omni-2B-RP | finetune | GGUF | 2.17B | M | other | ragraph-ai | 2 | 607 |
| JonusNattapong/lfm3-omni-dualhead | finetune | — | 75.0M | S | apache-2.0 | JonusNattapong | 1 | 635 |
| mradermacher/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | mradermacher | 0 | 677 |
| Ohjaaja/LFM3-Omni-2B-Q8_0_XL-GGUF | quantized | GGUF | — | — | other | Ohjaaja | 2 | 514 |
| Guilherme34/LFM3-Omni-2B-Looped-Latent-v2.1-GGUF | quantized | GGUF | — | — | — | Guilherme34 | 0 | 577 |
| MichaelAnthony/LFM3-Omni-2B-98K-V2-GGUF | quantized | GGUF | — | — | other | MichaelAnthony | 0 | 403 |
| postgrammar/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | postgrammar | 0 | 392 |
| BIGJUTT/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | BIGJUTT | 0 | 348 |
| Emerald7664/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | Emerald7664 | 0 | 327 |
| konnosena/LFM3-Omni-2B-JA-Mix-GGUF | quantized | GGUF | — | — | other | konnosena | 0 | 297 |
| brainworkup/LFM3-Omni-2B-MLX-oQ8e | quantized | MLX | 2.17B | M | — | brainworkup | 1 | 243 |
| NoxNotreve/LFM3-Omni-2B-GGUF | quantized | GGUF | — | — | other | NoxNotreve | 0 | 291 |
| fusorlabs/lfm3-omni-2b-web | finetune | — | — | — | other | fusorlabs | 1 | 181 |
| OsaurusAI/LFM3-Omni-2B-JANG_6M | quantized | — | — | — | other | OsaurusAI | 0 | 216 |
| abliter8-ai/lfm3-2b-omni-abliter8-mlx | finetune | mlx | — | — | other | abliter8-ai | 4 | 0 |
| cortex-ai-nova/NOVA-Judex-4.8-Omni-ft | quantized | GGUF | 2.17B | M | other | cortex-ai-nova | 0 | 181 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
Hub snapshot: 2026-11-15 · Providers: 2026-11-15 06:55 UTC · Articles: 2026-11-15 06:56 UTC · Summary: 2026-11-06 17:40 UTC · Last verified 2026-11-15 06:56 UTC
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
