flint-nvfp4-mlx is an open-weight model for text generation from CharizardQwQ, released under Apache License 2.0. It has 39.8B parameters. At 16-bit it needs about 95.5 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
Parameters39.8B
Context—
Weights—
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—
Runs On
What it takes to serve flint-nvfp4-mlx (39.8B 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 | 79.6 GB | 95.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x MI325X $2.00 · 1x MI355X $2.59 |
| 8-bit | 39.8 GB | 47.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 19.9 GB | 23.9 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 23, 2026.
flint-nvfp4-mlx on every accelerator the SAVRN Index prices, at every precision
Model Card
Identity and Version
- Repository
- klcharizardqwq/flint-nvfp4-mlx
- Publisher
- CharizardQwQ
- Task
- Text generation
- Modality
- Text
- Library
- mlx-lm
- Parameters
- 39.8B parameters
- Languages
- mlx-lm, mlx, moe
- Revision
- b3cce09ab7cb24dc79810b31d12674a6a0ed2952
- First published
- 2026-10-22
- Last updated
- 2026-10-22
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
Download from CharizardQwQ
Released by CharizardQwQ through its official repository on Hugging Face. Read the license.
Built From
- Derived from RecursiveSI/flint
- Quantized from RecursiveSI/flint
Memory Requirements
| Precision | Weights in memory |
|---|---|
| 16-bit | 79.6 GB |
| 8-bit | 39.8 GB |
| 4-bit | 19.9 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About flint-nvfp4-mlx
How much GPU memory does flint-nvfp4-mlx need?
About 95.5 GB at 16-bit and 23.9 GB at 4-bit: the weights (39.8B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run flint-nvfp4-mlx 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 flint-nvfp4-mlx commercially?
Yes. flint-nvfp4-mlx is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
Similar Models
One upstream MoE, six generations of masks, one bit-exact surgery, a rebuilt factual memory, and a three-stage post-training (SFT → SimPO → RLVR) — with every error I made and every idea I falsified along the way. This repository is the complete technical record of the project: the activation-instrumentation artifacts, the mask-generation methodology, the surgery code, the PLE (n-gram memory) rebuild, the trainers, and the full lab notebook. The champion weights live in sibling repos What I set out to do: take Qwen3.8-Flash-Next — an ~180B-class MoE (~130B of core network + 51B of n-gram memory table + 4B MTP) — and cut it by 75% of its experts (512 → 128) while conserving the maximum…
Research preview. main is now the healed checkpoint. It is measurably worse than the original (KLD 0.091, +0.058 nats) and loops occasionally in long generations (2 of 48, the original 0 of 48). Read "Known issues" first. IFM/K2-Horizon-MoVA-36B-A4B with its attention converted from grouped-query attention (GQA) to multi-head latent attention (MLA), DeepSeek-style, without retraining. The KV cache shrinks from 192 KiB to 84 KiB per token in BF16 (42 KiB with an 8-bit latent). The routed experts and the value experts are the original weights. As far as we know this is the first MLA conversion of a GQA mixture-of-value-experts model. It is a research preview: calibrated, then healed by…
This seedoss model was trained 2x faster with Unsloth and Huggingface's TRL library.
Chirp Chirp! We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement. Ornith-1.5 extends Ornith-1.0 (which was developed on top of Qwen3.5 and Gemma4 with additional continued pretraining, mid-training, and post-training) by expanding the self-improvement loop from scaffold and rollout optimization to jointly optimizing task generation, scaffold construction, and solution rollouts. Rather than relying on a fixed set of human-curated tasks and manually designed harnesses, Ornith-1.5 continuously generates new training tasks, discovers effective strategies for solving them, and improves the policy through reinforcement learning. For…
TemaQ-X7-Thinking（天馬求） は、Ornith AIが開発したモデル Ornith-1.5 を基盤にした、エージェント向けの大規模言語モデル（LLM）です。 TemaQ-X7-Thinking (TemaQ model) is an enhanced Large Language Model (LLM) designed for agent applications, based on the Ornith-1.5 model developed by Ornith AI. It is engineered to generate more flexible and useful responses, even for prompts that are difficult for standard models to handle effectively. When used in conjunction with TemaQ Agent, it enables advanced reasoning capabilities. ユーザーの責任: モデルの利用者は、生成されたコンテンツが、適用される法律、規制、およびHugging Faceの利用規約/コンテンツポリシーに準拠することを全面的に保証する必要があります。