NEWMEN
← back to all models

OpenAI: gpt-oss-20b

openai/gpt-oss-20b

OpenAIclosed-weight131K context14 providersIntelligence 62.0· est.

Cheapest provider

$0.03 / 1M

DekaLLM

Fastest provider (p95)

2705 tok/s

Groq

Intelligence (estimated)

62.0

Family + generation + popularity + price tier

Per-provider performance

Latency / throughput / uptime / price measured across providers over the last 30 minutes of live traffic. This is what proves “sourced cheapest” — Atlas mode draws on these per call to serve the cheapest path that holds quality.

ProviderQuantInput $/1MOutput $/1MLatency p50 / p95Throughput p50 / p95Uptime 30mSuccess
DekaLLMfull· bf16$0.0290$0.1400821ms / 7671ms71 / 123 tok/s90.93%90.6%
DeepInfrafull· bf16$0.0300$0.1400332ms / 6487ms57 / 92 tok/s99.76%99.8%
Novitaq4· fp4$0.0400$0.1500690ms / 4704ms133 / 166 tok/s99.95%100.0%
Parasailq4· fp4$0.0400$0.2000382ms / 2346ms131 / 218 tok/s99.47%99.4%
Phalafull· unknown$0.0400$0.1500429ms / 5210ms59 / 74 tok/s96.02%72.0%
SiliconFlowq8· fp8$0.0400$0.18001779ms / 6708ms51 / 151.1 tok/s99.24%99.8%
Togetherfull· unknown$0.0500$0.2000376ms / 5970ms108 / 135.6 tok/s83.05%80.2%
WandBq4· fp4$0.0500$0.20002365ms / 5340ms282 / 366 tok/s98.79%98.8%
Amazon Bedrockfull· unknown$0.0700$0.15002127ms / 14830ms150 / 381 tok/s67.8%
Amazon Bedrockfull· unknown$0.0700$0.15002127ms / 14830ms150 / 381 tok/s67.8%
Fireworksfull· unknown$0.0700$0.3000654ms / 3997ms80 / 161 tok/s98.68%98.6%
Googlefull· unknown$0.0700$0.2500317ms / 1256ms52 / 114 tok/s85.37%86.7%
Groqfull· unknown$0.0750$0.3000247ms / 2053ms393.5 / 2704.7 tok/s99.64%76.4%
NextBitq8· fp8$0.1000$0.4500766ms / 7038ms152 / 216.4 tok/s90.49%88.3%

“—” means live telemetry hasn’t accumulated enough recent traffic for that endpoint. “undisclosed” means the provider serves the model but doesn’t expose the quantization label (typically running fp8 / int8 internally).

Intelligence estimate

No public benchmark numbers indexed for this model yet, so the leaderboard score is derived from the catalogue data we sync: model family standing, generation, and live usage rank — applied identically to open- and closed-weight models. The heuristic ceiling sits below confirmed-benchmark frontier models so curated rankings stay clearly on top.

Want a curated score? File a benchmark report and we’ll add it to the next sync.

How Atlas mode sources OpenAI: gpt-oss-20b

  • Strict mode — pin OpenAI: gpt-oss-20b exactly and we pass it straight through, sourced from the cheapest provider above. The same model, no substitutions — currently DekaLLM at $0.03/1M.
  • Atlas mode — the default. Each call is auto-optimized for the cheapest path that holds quality, at least 5% off going direct from call one and climbing as it ramps. You always see which model served the call and exactly what you saved — thumbs-down anything you don’t like for a full refund.