What GPU do I need to run CohereLabs/cohere-transcribe-03-2026?
2.1B parameters, published in BF16. View on Hugging FaceGated
cohere-transcribe-03-2026 is published by CohereLabs on Hugging Face, with 409,109 downloads and 1,099 likes to date. It's a CohereAsrForConditionalGeneration model built for automatic-speech-recognition, published natively in BF16, and gated: you'll need to accept the model's terms on Hugging Face before downloading weights.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for activations and allocator fragmentation. Speech models don't build the same growing KV-cache a text model does. Memory scales primarily with input audio length. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 3.8 GB | 4.6 GB | RTX 3070 | 1 | $0.088/hr |
| FP8 (quantized) | 1.9 GB | 2.3 GB | RTX 5060 Ti | 1 | $0.145/hr |
| INT4 (quantized) | 1.0 GB | 1.2 GB | RTX 3070 | 1 | $0.088/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run cohere-transcribe-03-2026 at its published (BF16) precision: 1× RTX 3070, at $0.088/hr per GPU ($0.088/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
cohere-transcribe-03-2026: common questions
How much VRAM does cohere-transcribe-03-2026 need?
4.6 GB at BF16, 2.3 GB at FP8 (quantized), 1.2 GB at INT4 (quantized). All of those sit under 6 GB, the smallest capacity this page reasons about, so VRAM is not what limits where this model runs. The 3.8 GB of weights plus inference overhead is the whole requirement.
Do I need approval to download cohere-transcribe-03-2026?
Yes. CohereLabs gates this repository on Hugging Face, so you have to accept its terms with a signed-in account before the weights will download. It does not change the 4.6 GB the model needs once you have them.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More CohereLabs models
- cohere-transcribe-arabic-07-2026 (2.1B, BF16)
- North-Mini-Code-1.0 (30.5B, BF16)
Lineage
Quantized and fine-tuned versions
Alternatives at this size
Other models for automatic-speech-recognition within about a third of cohere-transcribe-03-2026's 2.1B parameters, from other model lines.
- whisper-large-v3 (1.5B, F16)
- Qwen3-ASR-1.7B (2.3B, BF16)
- seamless-m4t-v2-large (2.3B, F32)
- granite-speech-4.1-2b (2.3B, BF16)
- GLM-ASR-Nano-2512 (2.3B, BF16)
More on cohere-transcribe-03-2026
Fits on an 8 GB GPU at BF16: every model that fits in 8 GB.
Best speech-to-text models: how cohere-transcribe-03-2026 ranks against the rest.
Related reading: RTX 3070 pricing and specs, Whisper variants compared, and What AI inference is.
Related on Aquanode: GPU requirements for c4ai-command-r-v01 and GPU requirements for stt-2.6b-en-trfs.
Your own GPU cloud accounts
One console for the GPU clouds you already rent from. Your keys stay on your machine.
$20 a month, flat