Skip to content
aiqbench

Tencent · model passport

Tencent: Hy3

hy3
  • Open weights
  • Apache 2.0 licence
  • 298.8B params
  • 262,144 context
  • Released 2026-07-02(HF repo created)
7/27axes covered
Overall · LMArena
1,453#65 of 396
Best percentile
84.8Instruction following
Cohort standing
62.5median 59.6

What the data says

Benchmarked on 7 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 64.8 percentile points between its best and worst axis — 84.8 pct on Instruction following versus 20 pct on Agent Arena.

  • Instruction following84.8 pct1,447Verify →
  • Overall83.8 pct1,453Verify →
  • Hard prompts83.6 pct1,475Verify →
  • Coding82.9 pct1,501Verify →
  • WebDev Arena62.6 pct1,507Verify →
  • Factuality Arena57.6 pct1,445Verify →
  • Agent Arena20 pct-0.060Verify →
  • Vision ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document 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
  • LiveBenchNot 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

Repositorytencent/Hy3
Parameters298.8B
ArchitectureHYV3ForCausalLM
LicenceApache 2.0

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "tencent/Hy3"
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 →

Use it hosted

API model IDtencent/hy3
Context length262,144 tokens
Max output128,000 tokens
Input price$0.13 per 1M tokens
Cached input price$0.03 per 1M tokens
Output price$0.53 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningLogit biasMax completion tokensMax tokensMin-pPresence penaltyReasoningReasoning effortRepetition penaltyResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-kTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"tencent/hy3","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

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.

  • hy3 stands at 62.5 (raw mean 67.9) — measured on 7 of 27 axes.

    standingtemplateVerify →

  • hy3 stands at 62.5 versus a cohort median of 59.6 across 201 models released within 180 days — above 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 →