RA
Rhymes AI
Multimodal model · United States ·Oct 2024

Aria

Multimodal Open (Restricted)
24.9B
Parameters
6
Benchmarks

About Aria

Aria is an AI model developed by Rhymes AI, in the multimodal category, released in 2024, made available as an open-weights model with 24.9B params.

On this page you'll find Aria's full specifications, including 6 benchmark results. Review provider pricing and benchmark scores below, or compare Aria head-to-head with other multimodal models.

→ More models by Rhymes AI → All multimodal models → AI benchmarks leaderboard

Links

Announcement ↗ Hugging Face ↗

Run Aria locally — what to buy

Share this build ↗

Sized to what Aria actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Good
AMD Radeon RX 7900 XT
Consumer GPU · ~$899
~44.2 tok/s
Q4 · fits full context · up to 8k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Better
AMD Radeon RX 7900 XTX
Consumer GPU · ~$999
~53 tok/s
Q4 · fits full context · up to 8k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Best
NVIDIA GeForce RTX 5090
Consumer GPU · ~$1,999
~111.8 tok/s
Q4 · fits full context · up to 8k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →

Software support

·ExLlamaV2 ·KoboldCpp ·LM Studio ·MLX ·Ollama ·SGLang ·TensorRT-LLM ·llama.cpp ·vLLM ?KTransformers ?text-generation-webui

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Consumer GPU
14.1–111.8 tok/s
10 options →
Apple
17.6–47 tok/s
17 options →
Datacenter GPU
47.7–120+ tok/s
13 options →

Community

Would you run Aria again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

Aria benchmark scores▶
Compare on the benchmark table →
Benchmark Category Score Variant
MMLU general 73.3% 5-shot-from-hf-readme
MATH math 50.8% from-hf-readme
ChartQA multimodal 86.4% from-hf-readme
DocVQA multimodal 92.600 from-hf-readme
MMMU multimodal 54.9% from-hf-readme
MathVista multimodal 66.1% from-hf-readme

More models like Aria

AI
NeuroChess
· 72.3K
 RA
Allegro
Rhymes AI · 3B
 Mistral AI
Mistral OCR 4
Mistral AI
 Tencent Hunyuan
Hy-Embodied-0.5-VLA
Tencent Hunyuan
 Anthropic
Claude Mythos 5
Anthropic
 Anthropic
Claude Fable 5
Anthropic

Frequently asked questions

What is Aria? ▶

Aria is an AI model developed by Rhymes AI, in the multimodal category, released in 2024. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Aria? ▶

Aria was developed by Rhymes AI and released in 2024.

Is Aria open source or proprietary? ▶

Aria ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.

How much does Aria cost? ▶

Pricing for Aria depends on the provider. See the providers table on this page for the latest API rates.

How does Aria perform on benchmarks? ▶

Aria is benchmarked across 6 evaluations on GenAIList, including MMLU (73.3%). See the full benchmark table below and compare it with other models.

No videos yet. Know a good one? Add it below.

Know a great video about Aria?