ByteDance: Seed OSS 36B Instruct

Text input Text output Unavailable
Author's Description

Seed-OSS-36B-Instruct is a 36B-parameter instruction-tuned reasoning language model from ByteDance’s Seed team, released under Apache-2.0. The model is optimized for general instruction following with strong performance in reasoning, mathematics, coding, tool use/agentic workflows, and multilingual tasks, and is intended for international (i18n) use cases. It is not currently possible to control the reasoning effort.

Key Specifications
Cost
$$$$
Context
131K
Parameters
36B
Released
Sep 02, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Stop Frequency Penalty Include Reasoning Seed Min P Top Logprobs Top P Presence Penalty Temperature Logit Bias Reasoning Max Tokens Logprobs
Features

This model supports the following features:

Reasoning
Performance Summary

ByteDance's Seed OSS 36B Instruct model demonstrates moderate speed performance, ranking in the 21st percentile, and offers moderate pricing, placing it in the 37th percentile. The model exhibits strong reliability with an 85% success rate, indicating few technical issues. In terms of performance across benchmarks, Seed OSS 36B Instruct shows particular strength in General Knowledge (98.5% accuracy, 62nd percentile), Coding (91.8% accuracy, 77th percentile), and Instruction Following (67.0% accuracy, 76th percentile). It also performs well in Email Classification (98.0% accuracy, 63rd percentile) and Reasoning (68.0% accuracy, 54th percentile). However, the model struggles significantly with Mathematics, achieving only 27.5% accuracy (17th percentile), and shows a notable weakness in Hallucinations, with 76.0% accuracy (19th percentile), suggesting it frequently fails to acknowledge uncertainty. Its Ethics performance is also relatively low at 92.0% accuracy (25th percentile). Overall, its strengths lie in practical application areas like coding and instruction following, while complex mathematical reasoning and appropriately handling unknown information present challenges.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.16
Completion $0.65

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Chutes
Chutes | bytedance/seed-oss-36b-instruct 131K $0.16 / 1M tokens $0.65 / 1M tokens
Chutes
Chutes | bytedance/seed-oss-36b-instruct 131K $0.16 / 1M tokens $0.65 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by bytedance