gpt-oss-20b
Together AI · chat model
gpt-oss-20b is listed here as a chat model from Together AI. This page shows simple API pricing, token limits, and capability flags so you can compare it with similar options.
Input
$0.0500 / 1M tokens
Output
$0.2000 / 1M tokens
Cached input
N/A
Context
N/A
Quick read
Best for
Use this page when you need a fast view of cost, context size, and supported features before testing the model in your own workload.
Things to verify
Always check the provider page for discounts, cache pricing, region rules, and any model limits that may not appear in public metadata.
Pricing
| Item | Price |
|---|---|
| Input | $0.0500 / 1M tokens |
| Output | $0.2000 / 1M tokens |
| Embedding | $0.0500 / 1M tokens |
Token limits
Max input tokens
128.0K
Capabilities
| Capability | Supported |
|---|---|
| Vision | — |
| Function calling | ✅ |
| Parallel function calling | ✅ |
| Tool choice | ✅ |
| Prompt caching | — |
| Reasoning | — |
| Response schema | ✅ |
| System messages | — |
| Audio input | — |
| Audio output | — |
| Web search | — |
| PDF input | — |
| Video input | — |
Benchmarks
Most benchmark rows are attached to the base model family rather than this provider route. Open benchmark explorer
| Benchmark | Score | Metric | Scope | Checked | Source |
|---|---|---|---|---|---|
| SWE-bench Verified | 60.7% | accuracy (%) | Base model: gpt-oss-20b (gpt-oss-20b) | 2026-05-31 | Link |
| Codeforces | 2516 Elo | Elo | Base model: gpt-oss-20b (gpt-oss-20b) | 2026-05-31 | Link |
| AIME 2025 (no tools) | 92.5% | Accuracy (%) | Base model: gpt-oss (gpt-oss-120b) | 2026-05-31 | Link |
| AIME 2025 (no tools) | 91.7% | Accuracy (%) | Base model: gpt-oss (gpt-oss-20b) | 2026-05-31 | Link |
| GPQA Diamond | 80.1% | Accuracy (%) | Base model: gpt-oss (gpt-oss-120b) | 2026-05-31 | Link |
| GPQA Diamond | 71.5% | Accuracy (%) | Base model: gpt-oss (gpt-oss-20b) | 2026-05-31 | Link |
| MMLU | 90.0% | Accuracy (%) | Base model: gpt-oss (gpt-oss-120b) | 2026-05-31 | Link |
| MMLU | 85.3% | Accuracy (%) | Base model: gpt-oss (gpt-oss-20b) | 2026-05-31 | Link |
| SWE-bench Verified | 62.4% | Accuracy (%) | Base model: gpt-oss (gpt-oss-120b) | 2026-05-31 | Link |
| SWE-bench Verified | 60.7% | Accuracy (%) | Base model: gpt-oss (gpt-oss-20b) | 2026-05-31 | Link |
| Codeforces (with tools) | 2622 | Elo | Base model: gpt-oss (gpt-oss-120b) | 2026-05-31 | Link |
| Codeforces (with tools) | 2516 | Elo | Base model: gpt-oss (gpt-oss-20b) | 2026-05-31 | Link |
| HealthBench | 57.6% | Score (%) | Base model: gpt-oss (gpt-oss-120b) | 2026-05-31 | Link |
| HealthBench | 42.5% | Score (%) | Base model: gpt-oss (gpt-oss-20b) | 2026-05-31 | Link |
| Artificial Analysis Intelligence Index | 24.5 | score | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| Artificial Analysis Coding Index | 18.5 | score | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| Artificial Analysis Agentic Index | 27.6 | score | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| GPQA Diamond | 68.8% | accuracy | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| Humanity's Last Exam | 9.8% | accuracy | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| IFBench | 65.1% | accuracy | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| SciCode | 34.4% | accuracy | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| Terminal-Bench Hard | 10.6% | accuracy | Base model: gpt-oss (openai/gpt-oss-20b:free) | 2026-05-31 | Link |
| Artificial Analysis Intelligence Index | 33.3 | score | Base model: gpt-oss (openai/gpt-oss-120b) | 2026-05-31 | Link |
| Artificial Analysis Coding Index | 28.6 | score | Base model: gpt-oss (openai/gpt-oss-120b) | 2026-05-31 | Link |
| Artificial Analysis Agentic Index | 37.9 | score | Base model: gpt-oss (openai/gpt-oss-120b) | 2026-05-31 | Link |
Sources
| Source links | |
| Pricing data | LiteLLM model cost map |
| Synced at | 2026-05-28 |
Docs
| Official docs |
Similar models
This list is ranked by overall similarity. Use filters to emphasize the lens that matters most for the replacement you are making.
Comparing from
| Model | Cost | Input shape | Features | Context | Why it is close |
|---|---|---|---|---|---|
| gpt-oss-20b Together AI | In $0.0500 / 1M tokens Out $0.2000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Current model Reference row |
| Model | Cost | Input shape | Features | Context | Why it is close |
|---|---|---|---|---|---|
| Meta-Llama-3.1-8B-Instruct-Turbo Together AI | In $0.1800 / 1M tokens Out $0.1800 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 77% |
| gpt-oss-20b fireworks_ai | In $0.0500 / 1M tokens Out $0.2000 / 1M tokens | text
Output: text | Function callingTool choiceResponse schema | 131.1K | Text covered Overall 75% |
| qwen3-coder-30b-a3b-instruct novita | In $0.0700 / 1M tokens Out $0.2700 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | 32.8K | Text covered Overall 75% |
| devstral-small vercel_ai_gateway | In $0.0700 / 1M tokens Out $0.2800 / 1M tokens | text
Output: text | Function callingTool choiceResponse schema | 128.0K | Text covered Overall 73% |
| gpt-oss-120b azure_ai | In $0.1500 / 1M tokens Out $0.6000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | 131.1K | Text covered Overall 72% |
| devstral-small-2505 Mistral | In $0.1000 / 1M tokens Out $0.3000 / 1M tokens | text
Output: text | Function callingTool choiceResponse schema | 128.0K | Text covered Overall 71% |
| Qwen3-Next-80B-A3B-Instruct Together AI | In $0.1500 / 1M tokens Out $1.5000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 69% |
| Qwen3-Next-80B-A3B-Thinking Together AI | In $0.1500 / 1M tokens Out $1.5000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 69% |
| Mixtral-8x7B-Instruct-v0.1 Together AI | In $0.6000 / 1M tokens Out $0.6000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 69% |
| GLM-4.5-Air-FP8 Together AI | In $0.2000 / 1M tokens Out $1.1000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 69% |
| Llama-3.3-70B-Instruct-Turbo Together AI | In $0.8800 / 1M tokens Out $0.8800 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 68% |
| Qwen3-235B-A22B-Instruct-2507-tput Together AI | In $0.2000 / 1M tokens Out $6.0000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 68% |
| Qwen3-235B-A22B-Thinking-2507 Together AI | In $0.6500 / 1M tokens Out $3.0000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 66% |
| Qwen3-Coder-480B-A35B-Instruct-FP8 Together AI | In $2.0000 / 1M tokens Out $2.0000 / 1M tokens | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 66% |
| Qwen2.5-72B-Instruct-Turbo Together AI | In N/A Out N/A | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 65% |
| Qwen2.5-7B-Instruct-Turbo Together AI | In N/A Out N/A | text
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 65% |
| Llama-4-Scout-17B-16E-Instruct Together AI | In $0.1800 / 1M tokens Out $0.5900 / 1M tokens | text image
Output: text | Function callingParallel function callingTool choiceResponse schema | N/A | Same provider Overall 59% |
| qwen-turbo dashscope | In $0.0500 / 1M tokens Out $0.2000 / 1M tokens | text
Output: unknown | Function callingTool choice | 16.4K | Partial I/O overlap Overall 30% Missing text |
| qwen-turbo-2024-11-01 dashscope | In $0.0500 / 1M tokens Out $0.2000 / 1M tokens | text
Output: unknown | Function callingTool choice | 8.2K | Partial I/O overlap Overall 30% Missing text |
| qwen-turbo-2025-04-28 dashscope | In $0.0500 / 1M tokens Out $0.2000 / 1M tokens | text
Output: unknown | Function callingTool choice | 16.4K | Partial I/O overlap Overall 30% Missing text |
| qwen-turbo-latest dashscope | In $0.0500 / 1M tokens Out $0.2000 / 1M tokens | text
Output: unknown | Function callingTool choice | 16.4K | Partial I/O overlap Overall 30% Missing text |
| Qwen2.5-Coder-32B-Instruct nscale | In $0.0600 / 1M tokens Out $0.2000 / 1M tokens | text
Output: unknown | Low overlap | N/A | Partial I/O overlap Overall 19% Missing text |
No models match this filter.