Qwen
Qwen
Language model · China ·Jun 2024

Qwen2-57B-A14B

Language Open (Restricted)
57B
Parameters
16
Benchmarks

About Qwen2-57B-A14B

Qwen2-57B-A14B is an AI model developed by Qwen, in the language category, released in 2024, made available as an open-weights model with 57B params.

On this page you'll find Qwen2-57B-A14B's full specifications, including 16 benchmark results. Review provider pricing and benchmark scores below, or compare Qwen2-57B-A14B head-to-head with other language models.

→ More models by Qwen → All language models → AI benchmarks leaderboard

Links

Announcement ↗ Hugging Face ↗

Run Qwen2-57B-A14B locally — what to buy

Share this build ↗

Sized to what Qwen2-57B-A14B 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
Apple M1 Max 64GB
Apple · ~$2,600
~10.3 tok/s
Q4 · fits full context · up to 8k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Better
Apple M2 Max 96GB
Apple · ~$3,300
~10.3 tok/s
Q4 · fits full context · up to 8k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Best
Apple M1 Ultra 64GB
Apple · ~$3,700
~20.6 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
6.2–7.4 tok/s
2 options →
Apple
10.3–20.6 tok/s
13 options →
Datacenter GPU
20.8–120+ tok/s
13 options →

Community

Would you run Qwen2-57B-A14B 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.

Qwen2-57B-A14B benchmark scores▶
Compare on the benchmark table →
Benchmark Category Score Variant
HumanEval code 53% base-from-hf-readme
MBPP code 71.9% base-from-hf-readme
EvalPlus coding 57.2 base-from-hf-readme
MultiPL-E coding 49.8 base-from-hf-readme
C-Eval general 87.7% base-from-hf-readme
MMLU general 76.5% base-from-hf-readme
MMLU-Pro general 43% base-from-hf-readme
ARC-Challenge knowledge 64.1 base-from-hf-readme
CMMLU knowledge 88.5 base-from-hf-readme
GSM8K math 80.7% base-from-hf-readme
MATH math 43% base-from-hf-readme
BIG-Bench Hard reasoning 67% base-from-hf-readme
GPQA Diamond reasoning 34.3% base-from-hf-readme
HellaSwag reasoning 85.2 base-from-hf-readme
Winogrande reasoning 79.5 base-from-hf-readme
TruthfulQA safety 57.7% base-from-hf-readme

More models like Qwen2-57B-A14B

AI
DMind-1
 AI
COLLECTIVE-1
 AI
Cogito 1 series
 AI
Euler (Euler大模型)
· 2.5B
 AI
SmolLM-1.7B
· 1.7B
 AI
Llama-3-70B-Synthia-v3.5
· 70B

Frequently asked questions

What is Qwen2-57B-A14B? ▶

Qwen2-57B-A14B is an AI model developed by Qwen, in the language category, released in 2024. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Qwen2-57B-A14B? ▶

Qwen2-57B-A14B was developed by Qwen and released in 2024.

Is Qwen2-57B-A14B open source or proprietary? ▶

Qwen2-57B-A14B 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 Qwen2-57B-A14B cost? ▶

Pricing for Qwen2-57B-A14B depends on the provider. See the providers table on this page for the latest API rates.

How does Qwen2-57B-A14B perform on benchmarks? ▶

Qwen2-57B-A14B is benchmarked across 16 evaluations on GenAIList, including HumanEval (53%). 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 Qwen2-57B-A14B?