AI NEWS.ccAI는 지금, 미래는 업데이트 중
♡ 저장한 도구 0
AI NEWS · 모델 정보기능을 이해하고 근거 있게 선택하세요. ↗기본 정보 우선 등록 · 상세 정보 지속 보완
THE AI MODEL DIRECTORY 모델 탐색을 시작하세요

나에게 맞는 AI 모델을 찾아보세요.

언어부터 이미지, 오픈 웨이트부터 API까지. 기능과 실제 비용을 비교하세요.
다음 아이디어에 맞는 AI를 찾아보세요.

인기 검색
85,003 개 모델13202 개 제공사 / 게시자45 개 기능 분류465개 비교 가능 요금 · 30개 공개 평가
T텍스트 생성▧이미지 생성▷영상 생성⌘코드 생성◖음성 합성♫음악 생성⬡3D 모델링▤문서 분석
DISCOVER & COMPARE

AI 모델 탐색 85003

주목할 모델을 찾아보세요

This model is a fine-tuned version of LeBenchmark/wav2vec2-FR-7K-large, trained on a composite dataset comprising of over 2200 hours of French speech audio, using the train and validation splits of Common Voice 11.0, Multilingual LibriSpeech, Voxpopuli, Multilingual TEDx, MediaSpeech, and African Accented French. When using the model make sure that your speech input is also sampled at 16Khz.

번역 대기 중 · 원문 표시

음성 인식
입력 오디오→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗

This is an experimental, refusal-ablated GGUF derivative of XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B. It deliberately reduces refusal on harmful prompts. Its responses require independent review and appropriate controls in any downstream application. **Evaluation status:** The judged safety evaluation is still running. The tables below report completed Q8_0 runs only; judged results will be added when complete. This is an edited derivative that was then quantized, not a plain quantization of the original weights. Text generation is the only evaluated modality; no vision capability is claimed for these GGUF files.

번역 대기 중 · 원문 표시

텍스트 생성
입력 텍스트→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗

Refusal removed from GLM-5.3-Flash by editing **666 of its 12,384 routed experts** — about **1.74% of the model's parameters**. Quantized to IQ1_S at **71.5 GiB**, which fits on a single 96 GB card. I did the edit and the quantization myself, and I measured what it cost. Every number below is first-party, taken on my own rig, on the file in this repo. > **Read this before you run it.** Every headline number below was measured at > **temperature 0, seed 0, repetition penalty 1.1/256, 16k context, `--reasoning-budget 1024`.** > The file itself ships `temperature 1.0` in its metadata and no reasoning budget, so a plain > `llama-server -m …` does **not** run the measured configuration. I tested that gap directly > (table under *Default settings vs tested settings*): at the file's own defaults, long coding > tasks ran out of output budget without answering about **1 time in 8**; with > `--reasoning-budget 1024` they did not. **For code and chat, use the quickstart line as > written. For hard math, leave the budget off** (see the correction below). This is an experimental 1.92-bit build: 37 of > 120 harmful-set responses and 8 of 120 benign ones hit the output cap in the headline run, and > that is reported separately from "unusable" below.

번역 대기 중 · 원문 표시

시각 질의응답
입력 텍스트 · 이미지→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗

An uncensored edit of **NVIDIA Nemotron-3.5-Bolt-74B-A9B**, a hybrid Mamba/MoE model, produced by editing NVIDIA's **shipped 24GB 2-bit GGUF** (nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB-GGUF) **directly** — no BF16 requantize at any point. That matters for this model specifically. Bolt is a LatentMoE: every routed expert sums through one dense `ffn_latent_up` per layer, and the experts themselves are ~17GB of IQ2_XXS weights produced by NVIDIA's quantization-aware distillation. A conventional abliterate-then-requantize pipeline would throw that QAD work away. Instead the projection is applied in place and codes are re-picked under the **original block scales**, so at λ=0 the output file is byte-identical to the input by construction — which is the self-test the build gates on (verified: all edited tensors byte-identical at λ=0). The published file is the same length as NVIDIA's, 22,018,326,528 bytes. Bolt is multimodal, so the projector ships alongside. The edit touched only text-model tensors, so the projector is byte-for-byte NVIDIA's file; download both or vision silently does nothing.

번역 대기 중 · 원문 표시

시각 질의응답
입력 텍스트 · 이미지→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗

GGUF quantizations of **orcarouter/GLM-5.3-Flash-Uncensored-FP8**, built with a custom importance matrix and measured rather than assumed. > ⚠️ **Uncensored model.** Safety refusals were removed by OrcaRouter. It will comply with > harmful, unethical, and illegal requests. Released for research — red-teaming, refusal- > mechanism study, robustness evaluation. You are responsible for what you do with it. > 🔧 **What happened, and where it stands (2026-09-06).** On 2026-09-05 a user reported > looping and incoherent output from `IQ2_S` (discussion #1). They were > right. The original build pinned only the `blk.45` MTP block at high precision; attention, > the **shared** experts, the three dense FFN blocks, and the embeddings were all quantized > down with the routed experts. The shared experts and attention run on every token. > > **`IQ2_S` was rebuilt on 2026-09-06** with those tensors held at `q8_0` (+7 GB) and tested > against the original on a rented H200: 12 tasks × 2 seeds × 2 samplers, same scorer, same > settings. Full table in **Coherence gate** below. Short version: the original failed 3 of 8 > coding tasks by reasoning to the output cap without ever answering; the rebuild passed > all 48, matching `IQ3_XXS`, which also passed all 48 on its original recipe. > **`IQ2_XXS` is being rebuilt the same way** and stays flagged until it passes the same gate. > > **Two corrections to earlier versions of this page.** The imatrix has **10 chunks, not 60** > (the run checkpointed at 10 and stopped; numerically sound, one uncovered expert row in > `blk.29`). And the repetition-penalty workaround I posted **did not change the outcome** on > this failure mode — 21/24 → 21/24 for the original file with and without it. It helps > n-gram repetition; it does not help a model that cannot emit the literal it intends. > The rebuilt files carry `repeat_penalty 1.1` in their metadata anyway, because it is cheap. > > Only `IQ3_XXS` had been benchmarked before publication, and I shipped all three off that. > That was the actual mistake. The gate below is the fix for the process, not just the file.

번역 대기 중 · 원문 표시

텍스트 생성
입력 텍스트→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗

> **v2 is out:** Ternary-Bonsai-2-27B-Abliterated-v2-PQ2_0-MTP-GGUF. > It fixes the thinking-mode problem described below: blank answers on harmful prompts drop > from 27% to 15% at default settings and to 0% with `reasoning_effort=medium`. This repo > stays up unchanged as v1. Refusal behaviour removed from `prism-ml/Ternary-Bonsai-2-27B-gguf` at PQ2_0 (2.13 bpw), by flipping a small number of ternary digits in place. **Block scales are never touched and the model is never requantised**, so this is the released ternary packing with a few digits changed — not a re-derived one. A second file adds the Qwen3.8-27B multi-token-prediction head, which makes `--spec-type draft-mtp` work and decodes **~37% faster** at identical weights.

번역 대기 중 · 원문 표시

텍스트 생성
입력 텍스트→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗

> **v2 is out:** v2, PTQ1_0 > fixes the thinking-mode problem described below and is the one to use with thinking on. > This v1 file stays up unchanged. Refusal-ablated **Ternary Bonsai 2 27B**, edited directly on the ternary lattice and shipped in PrismML's native PTQ1_0 pack. - **5,946,648,928 bytes — byte-for-byte the size of the official PTQ1_0 release** - No requantization, no BF16 bake, no runtime steering, no control vector at serve time - Requires the PrismML llama.cpp fork

번역 대기 중 · 원문 표시

텍스트 생성
입력 텍스트→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗

**v2 is built to be run with thinking on.** Abliterated Bonsai 2 builds, my own v1 included, tend to write the whole answer inside their reasoning and run out of budget before handing it over. v2 changes one row of the output layer so the model closes its reasoning once the answer is ready. With `reasoning_effort` set to medium it answered every prompt on the main test; at default settings it halves v1's blank answers. **Where v2 stands out: hard requests, given room to think.** StrongReject (150 detailed harmful requests), thinking on, default settings, 16,384-token budget, same judge for every row: Much of that gap is answers the other builds never deliver: they write the answer inside their reasoning and run out of room before closing it.

번역 대기 중 · 원문 표시

텍스트 생성
입력 텍스트→ 출력 텍스트
요금 미확인멀티모달 및 채널 요금 별도 확인
자세히 보기 ↗
출처 / 문서 ↗
COST EXPLORER

좋은 모델, 명확한 비용.

비용 분석

월간 토큰 사용량으로 예상 API 비용을 비교하세요.

1 M = 100만 토큰 · USD
동일 채널의 기본 텍스트 토큰 요금 비교

수집된 백만 토큰당 USD 요금입니다. 캐시 할인, 구간 요금, 추가 추론 출력, 세금은 제외됩니다. 최신 요금은 출처에서 확인하세요.

MODEL HANDBOOK

이해하고 선택하세요

선택 가이드 및 FAQ
API 요금은 어떻게 비교하나요?+

입력, 출력, 캐시 요금을 구분하고 동일 작업량으로 비교하세요. 이미지, 음성, 영상은 개수, 시간, 해상도로 과금될 수 있습니다.

오픈 웨이트는 무료 상업 이용을 뜻하나요?+

반드시 그렇지는 않습니다. 버전별 라이선스, 상업 이용 제한, 정책을 확인하세요. 자체 호스팅에는 연산 및 유지 비용도 듭니다.

1위 모델이 나에게 가장 적합한가요?+

순위는 작업, 버전, 평가 방식에 따라 다릅니다. 실제 업무로 정확도, 지연, 안정성, 비용을 비교하세요.

AI NEWS · 정보에서 선택으로

뉴스로 흐름을, 도구로 효율을, 모델로 가능성을.

AI 뉴스 탐색 ↗AI 도구 찾기 ↗