Скорость вывода
3319.9 tok/s
4258.1 tok/s
6GPT-OSS243.2 tok/s
7229.0 tok/s
GPT-OSS is available through 200 API providers on LMSpeed. Compare API pricing from $0.0010 to $1498.50 per million input tokens across providers. Free API access is offered by 11 providers. Context window: 131,072. In speed benchmarks, the fastest provider reaches 1796 tok/s.
GPT-OSS is an open-weight language model family designed for self-hosted inference, research, and cost-efficient alternatives to proprietary GPT-class models.
Также известна как
Compare GPT-OSS API pricing across 189 providers. Input prices range from $0.0010 to $1498.50 per million input. Xiao Wan offers the lowest rate at $0.0010/M. 11 providers offer free API credits or a free tier.
| Провайдер | Вариант модели | Аудит | Входные данные ($/M) | Выходные данные ($/M) | Скорость (т/с) | Первый токен |
|---|---|---|---|---|---|---|
VSLLM99.7% | gpt-oss-120b | — | $4.38 | $13.14 | 1319.0 t/s+442% | 0.61 s |
钠 API100% | gpt-oss-120b-medium | — | $0.055-9% | $0.110-45% | 255.8 t/s+5% | 1.45 s |
素墨API100% | openai/gpt-oss-20b | — | $0.010-83% | $0.010-95% | 234.4 t/s | 1.48 s |
| openai/gpt-oss-120b | — | $0.010-83% | $0.010-95% | — | — | |
| openai/gpt-oss-120b:free | — | $0.010-83% | $0.010-95% | — | — | |
| openai/gpt-oss-20b:free | — | $0.010-83% | $0.010-95% | — | — | |
小水管 API100% | gpt-oss-20b | 648468100 | Бесплатно | Бесплатно | 216.3 t/s | 1.31 s |
| gpt-oss-120b | — | Бесплатно | Бесплатно | — | — | |
星见雅 API100% | 英伟达/openai/gpt-oss-120b | — | Бесплатно | Бесплатно | 144.0 t/s | 0.94 s |
| openai/gpt-oss-120b | — | Бесплатно | Бесплатно | 142.7 t/s | 1.29 s | |
| 英伟达/openai/gpt-oss-20b | — | Бесплатно | Бесплатно | — | — | |
6345ywz API99.8% | gpt-oss-120b | — | $0.137 | $0.575 | 126.7 t/s | 0.79 s |
| openai/gpt-oss-20b | — | $0.0041-93% | $0.022-89% | — | — | |
| openai/gpt-oss-120b | — | $0.021-66% | $0.082-59% | — | — | |
| gpt-oss-20b | — | $0.055-9% | $0.247 | — | — | |
WSocket AI99.3% | openai/gpt-oss-120b | 8484100100 | $1498.50 | $5994.00 | 124.7 t/s | 2.30 s |
| openai/gpt-oss-20b | — | $499.50 | $1998.00 | — | — | |
V-API100% | gpt-oss-120b | — | $0.018-70% | $0.018-91% | — | — |
老张API100% | gpt-oss-20b | — | $0.096 | $0.384 | — | — |
| gpt-oss-120b | — | $0.479 | $1.92 | — | — |
Данные о ценах из публичных API провайдеров
Сравните производительность по скорости и задержке у всех API-провайдеров.
| Провайдер | Скорость | Задержка | Тесты |
|---|---|---|---|
玄黄 gpt-oss-120b | 1796.31 tok/s | 0.49s | 5 |
gpt-oss-120b | 1677.82 tok/s | 0.56s | 10 |
gpt-oss-120b | 1637.28 tok/s | 0.91s | 10 |
gpt-oss-120b | 1467.36 tok/s | 0.82s | 5 |
gpt-oss-120b | 1319.02 tok/s | 0.61s | 10 |
gpt-oss-120b | 970.77 tok/s | 0.94s | 5 |
gpt-oss-120b | 516.58 tok/s | 1.41s | 75 |
openai/gpt-oss-120b | 481.90 tok/s | 0.43s | 5 |
accounts/fireworks/models/gpt-oss-20b | 359.18 tok/s | 1.14s | 10 |
gpt-oss-20b:free | 339.45 tok/s | 2.51s | 5 |
gpt-oss-120b-medium | 337.69 tok/s | 2.32s | 20 |
gpt-oss-120b:free | 257.64 tok/s | 2.50s | 5 |
gpt-oss-120b-medium | 255.84 tok/s | 1.45s | 5 |
gpt-oss-20b | 246.32 tok/s | 2.13s | 20 |
gpt-oss-120b-medium | 244.47 tok/s | 1.16s | 5 |
gpt-oss:20b | 243.42 tok/s | 2.62s | 5 |
openai/gpt-oss-120b:novita | 240.32 tok/s | 1.26s | 5 |
openai/gpt-oss-20b | 234.37 tok/s | 1.48s | 15 |
openai/gpt-oss-20b | 227.95 tok/s | 1.47s | 5 |
openai/gpt-oss-120b | 217.84 tok/s | 7.42s | 60 |
Показано 1–20 из 39 провайдеров
deepseek-v3-2
DeepSeek V3.2 is an upgraded V3-series MoE model with stronger reasoning, coding, and math performance, widely available through OpenAI-compatible API relays.
gpt-5-4
OpenAI GPT-5.4 extends the GPT-5 family with stronger instruction following, deeper tool use, and improved performance on coding, math, and long-document analysis.
kimi-k2-5
Moonshot Kimi K2.5 is an open-weight multimodal agent model with native vision and text input, strong coding performance, and a 256K context window.
gpt-5-3-codex
OpenAI GPT-5.3 Codex is a code-specialized variant in the GPT-5 series, optimized for code generation, debugging, and software development tasks.
minimax-m2-5
MiniMax M2.5 is MiniMax's flagship text model for coding and agents, with SOTA-level programming and agentic performance, improved token efficiency, and fast high-TPS API deployment.
deepseek-v4-flash
DeepSeek V4 Flash is a fast, cost-efficient language model in the DeepSeek V4 family, optimized for low-latency chat, coding assistance, and high-throughput API workloads while retaining strong reasoning quality.
Рейтинги основаны на тестах, предоставленных сообществом, и периодических зондах работоспособности. Носит рекомендательный характер, не является официальными данными.Artificial Analysis
