MODEL PROOF
gpt-oss-20b
FRONTIER
Nothing is both better and cheaper under this workload.
Serving precision differs between offers.
- Independent LMArena score
- 1,287.7 ± 6.4 · 10,368 votes
- Context window
- 131.07K
- Maximum output
- 32.77K
- Input modalities
- text
- Output modalities
- text
- Published input price
- $0.018 / 1M tokens
- Published output price
- $0.090 / 1M tokens
- Pricing kind
- fixed
Inspect complete billing conditions and endpoint terms below.
Price from Darkbloom: the cheapest offer that is serving, at standard delivery and at a declared precision that is not 4-bit (fp8), at its standard rate. The maker’s own precision is not known.
Every price dimension · USD per million tokens
| Input | $0.018/M |
|---|---|
| Output | $0.090/M |
| Cached input | $0.0090/M |
| Cache write | Unknown/M |
| Cache write 1h | Unknown/M |
| Reasoning | Unknown/M |
Who sells it
The same weights, different shops. Cheapest is not like-for-like when serving precision differs.
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.029/M from DekaLLM.
12 provider offers · rates, limits and conditions
Rates are USD per million tokens. Endpoint terms can differ even when the quoted price and precision match. A listed parameter is a provider declaration, not a task-success test.
| Seller | Input | Output | Precision | Uptime (1d) |
|---|---|---|---|---|
| Darkbloomdarkbloom/fp8 Endpoint terms
| $0.018 | $0.090 | fp8 | 99.72% |
| AkashMLakashml/fp4 Endpoint terms
| $0.020 | $0.100 | fp4 | 99.19% |
| DekaLLMdekallm/bf16 Endpoint terms
| $0.029 | $0.140 | bf16 | 99.45% |
| DeepInfradeepinfra/bf16 Endpoint terms
| $0.030 | $0.140 | bf16 | 99.92% |
| CoreWeavecoreweave/fp4 Endpoint terms
| $0.030 | $0.130 | fp4 | 99.94% |
| Parasailparasail/fp4 Endpoint terms
| $0.030 | $0.150 | fp4 | 99.92% |
| SiliconFlowsiliconflow/fp8 Endpoint terms
| $0.040 | $0.180 | fp8 | 84.84% |
| Novitanovita/fp4 Endpoint terms
| $0.040 | $0.150 | fp4 | 99.96% |
| Amazon Bedrockamazon-bedrock/eu-west-1 Endpoint terms
| $0.070 | $0.150 | undeclared | 99.94% |
| Amazon Bedrockamazon-bedrock Endpoint terms
| $0.070 | $0.150 | undeclared | 99.95% |
| Googlegoogle-vertex/us-central1 Endpoint terms
| $0.070 | $0.250 | undeclared | 96.42% |
| Groqgroq Endpoint terms
| $0.075 | $0.300 | undeclared | 97.91% |
Independent scores
| LMArena Elo | 1287.7default |
|---|
LMArena Elo under CC BY 4.0. Absence of another board is not a score of zero.
Best rankings by task
- Data visualisation #115 940
- Websites #129 859
Per-task Elo and rank from Design Arena, via OpenRouter’s model feed. The rank is the feed’s, and it counts models OpenRouter does not list.
Capability
| Context window | 131K |
|---|---|
| Max output | 32K |
| Input modes | text |
| Tool use | yes |
| Reasoning | always on |
| Knowledge cutoff | 2024-06-30 |
| Open weights | Unknown |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-10-06 |
| Quality data | verified |
gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...
Questions this page answers
What does gpt-oss-20b cost?
$0.018/M in, $0.090/M out in this catalogue, as of the fetch date on this page. That is the API row, not a subscription.
Does gpt-oss-20b have an independent quality score?
gpt-oss-20b has an LMArena score in this catalogue.
What beats gpt-oss-20b?
Nothing is both better and cheaper than gpt-oss-20b.
Does this page use Artificial Analysis scores for gpt-oss-20b?
No. Artificial Analysis figures are not published here. Quality on this page is LMArena Elo where a score exists; otherwise the row is unrated.
Is the cheapest gpt-oss-20b endpoint the same product?
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.029/M from DekaLLM.
MODEL MONUMENT
Save or share this proof
Badge
The verdict as one SVG, for a README or a docs page. It states this model's dominance status at the balanced workload on the LMArena lens, with the date it was computed, and it is rebuilt with the catalogue — so it changes when the verdict changes, including to one you would rather it did not.
[](https://undominated.ai/models/openai__gpt-oss-20b/)<a href="https://undominated.ai/models/openai__gpt-oss-20b/"><img src="https://undominated.ai/badge/openai__gpt-oss-20b.svg" alt="gpt-oss-20b — Undominated.ai dominance verdict" height="36"></a>Direct file: https://undominated.ai/badge/openai__gpt-oss-20b.svg — a static SVG, written by scripts/build-badges.mjs on every build, so the copy you embed is never
older than the last deploy.
