Skip to content
aiqbench

Hugging Face · model passport

Zephyr 7b Alpha

zephyr-7b-alpha
  • Open weights
  • MIT licence
  • 7.2B params
  • Released 2023-10-11(curated)
5/27axes covered
Overall · LMArena
1,127#372 of 396
Best percentile
8.2Coding
Cohort standing
22.9median 26.9

What the data says

Benchmarked on 5 of 27 tracked axes. Every figure below is the model's own published result on that board.

Weakest axes

  • 6.2 pct — LiveBench (18.6%)
  • 6.3 pct — Hard prompts (1,123)
  • 6.3 pct — Instruction following (1,099)

Uneven: 2 percentile points between its best and worst axis — 8.2 pct on Coding versus 6.2 pct on LiveBench.

  • Coding8.2 pct1,167Verify →
  • Overall6.3 pct1,127Verify →
  • Instruction following6.3 pct1,099Verify →
  • Hard prompts6.3 pct1,123Verify →
  • LiveBench6.2 pct18.6%Verify →
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Run it yourself

RepositoryHuggingFaceH4/zephyr-7b-alpha
Parameters7.2B
ArchitectureMistralForCausalLM
LicenceMIT

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "HuggingFaceH4/zephyr-7b-alpha"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • zephyr-7b-alpha stands at 22.9 (raw mean 6.7) — measured on 5 of 27 axes.

    standingtemplateVerify →

  • zephyr-7b-alpha stands at 22.9 versus a cohort median of 26.9 across 65 models released within 180 days — below the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →