Compare / GPT-6 Astra vs gpt-oss-120b
GPT-6 Astra vs gpt-oss-120b
Which is cheaper, which is more capable, and which does better at each kind of work, from 25 results on 6 sources that measured both models.
Of the 25 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee).
GPT-6 Astra or gpt-oss-120b? The short answer
Pick GPT-6 Astra if quality matters most: it scores higher (Intelligence Index 52.7 against 11.6). Pick gpt-oss-120b if cost matters more: it costs 100 times less per token. gpt-oss-120b is also faster.
- gpt-oss-120b is cheaper: $0.20 against $20.00 per million tokens, 100 times less.
- GPT-6 Astra scores higher on the Artificial Analysis Intelligence Index: 52.7 against 11.6.
- gpt-oss-120b writes faster: 259 tokens a second against 72.
- GPT-6 Astra is newer: released 3 Sep 2026, against 5 Aug 2025 for gpt-oss-120b.
- On third-party benchmarks, on the results both models have: GPT-6 Astra does better for agents and tool use, professional work, reasoning and knowledge, writing and sticking to the facts.
Head to head
Every fact side by side
| GPT-6 Astra | gpt-oss-120b | |
|---|---|---|
| Price per million tokens | $10.00 in · $50.00 out | $0.10 in · $0.50 out |
| Blended (3 in : 1 out) | $20.00 | $0.20 |
| Cheapest host, blended | $20.00 (Amazon Bedrock) | $0.060 (Near AI, FP4) |
| Cached input, per million | $1.00 cached input | – |
| Intelligence Index | 52.7 | 11.6 |
| Spring Prompt overall | 77 | – |
| Output speed, tokens a second | 72 | 259 |
| Context window | 1,050,000 tokens | 131,072 tokens |
| Released | 3 Sep 2026 (newer) | 5 Aug 2025 |
| Weights | Closed | Open |
| Developer based in | the US | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
From our research
- Best AI for Product Descriptions: Accuracy, Cost and Real Examples 20 Oct 2026
- Qwen4-Max review: top of our slide deck ratings, weak on product listings 16 Nov 2026
- Claude Fable 5.2 review: first on our survey and ad budget tests, and one of the dearest models to run 10 Nov 2026
- GPT-6.1 Terra review: close to the top of our business tests at half the price of Sol 5 Nov 2026
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | GPT-6 Astra | gpt-oss-120b |
|---|---|---|
| APEX-Agents reported by APEX-Agents · 1 result | ||
| Tasks passed% of tasks, higher is better 7 · 37 of 37 measured | 64.7% | 4.4% |
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 26 · 187 of 387 measured | 1,477max reasoning | 1,352 |
| Overall · Text factualityArena rating, higher is better 31 · 142 of 163 measured | 1,466max reasoning | 1,385 |
| Business, management and finance · TextArena rating, higher is better 29 · 183 of 380 measured | 1,473max reasoning | 1,352 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 4 · 162 of 249 measured | 52.7max reasoning | 11.6high reasoning |
| GPQA Diamond% of questions, higher is better 1 · 119 of 232 measured | 96.3%xhigh reasoning | 78.2%high reasoning |
| Humanity's Last Exam% of questions, higher is better 7 · 118 of 250 measured | 54.7%max reasoning | 19.6%high reasoning |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 13 · 116 of 230 measured | 68.9high reasoning | 38.5medium reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 1 · 224 of 230 measured | 0.0%high reasoning | 107.0%medium reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 35 · 217 of 230 measured | 0.38medium reasoning | 0.29medium reasoning |
| Vectara Hallucination Leaderboard reported by Vectara Hallucination Leaderboard · 2 results | ||
| Hallucination rate · Vectara% of summaries, lower is better 45 · 88 of 104 measured | 8.7% | 14.2% |
| Answer rate · Vectara% of documents, higher is better 1 · 17 of 104 measured | 100.0% | 99.9% |
| Vending-Bench 2 reported by Vending-Bench 2 · 1 result | ||
| Money after a yearUS dollars, higher is better 1 · 56 of 58 measured | $15,515 | $-21.53 |
Show all 25 results (12 more)
| Measure | GPT-6 Astra | gpt-oss-120b |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results | ||
| Creative writing · TextArena rating, higher is better 32 · 238 of 385 measured | 1,449max reasoning | 1,278 |
| Expert prompts · TextArena rating, higher is better 17 · 188 of 338 measured | 1,517max reasoning | 1,357 |
| Instruction following · TextArena rating, higher is better 24 · 204 of 387 measured | 1,474max reasoning | 1,323 |
| Writing, literature and language · TextArena rating, higher is better 24 · 215 of 386 measured | 1,465max reasoning | 1,309 |
| Artificial Analysis reported by Artificial Analysis · 8 results | ||
| Terminal-Bench 2.1% of tasks, higher is better 2 · 88 of 131 measured | 89.9%high reasoning | 26.2%high reasoning |
| Terminal-Bench 4.0% of tasks, higher is better 3 · 72 of 120 measured | 59.6%xhigh reasoning | 0.0%high reasoning |
| τ-bench banking% of tasks, higher is better 15 · 67 of 121 measured | 43.1%xhigh reasoning | 12.8%high reasoning |
| Artificial Analysis Coding Indexindex score, higher is better 4 · 85 of 130 measured | 77.1high reasoning | 30.4high reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 36 · 160 of 241 measured | 80.7%max reasoning | 52.0%high reasoning |
| Output speedtokens per second, higher is better 59 · 8 of 78 measured | 71.7max reasoning | 259high reasoning |
| SciCode% of problems, higher is better 21 · 101 of 121 measured | 56.5%max reasoning | 34.0%high reasoning |
| Time to first answer tokenseconds, lower is better 33 · 45 of 78 measured | 1.41low reasoning | 8.22high reasoning |
Compare them with other models
Quick answers
Is GPT-6 Astra better than gpt-oss-120b?
Pick GPT-6 Astra if quality matters most: it scores higher (Intelligence Index 52.7 against 11.6). Pick gpt-oss-120b if cost matters more: it costs 100 times less per token. gpt-oss-120b is also faster.
Which is cheaper, GPT-6 Astra or gpt-oss-120b?
gpt-oss-120b is cheaper: $0.20 against $20.00 per million tokens, 100 times less.
Which is faster, GPT-6 Astra or gpt-oss-120b?
gpt-oss-120b writes about 259 tokens a second on its usual API, against 72 for GPT-6 Astra (Artificial Analysis).
Which is better for agents and tool use, GPT-6 Astra or gpt-oss-120b?
GPT-6 Astra ranks 4 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 125 for gpt-oss-120b.
Which is better for professional work, GPT-6 Astra or gpt-oss-120b?
GPT-6 Astra ranks 5 of 51 models on sale for real tasks from banking, consulting and law, business documents and freelance projects, against 49 for gpt-oss-120b.
Which is better for reasoning and knowledge, GPT-6 Astra or gpt-oss-120b?
GPT-6 Astra ranks 3 of 187 models on sale for hard questions across science, maths and general knowledge, against 118 for gpt-oss-120b.
Which is better for writing, GPT-6 Astra or gpt-oss-120b?
GPT-6 Astra ranks 20 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 124 for gpt-oss-120b.
Which has the bigger context window, GPT-6 Astra or gpt-oss-120b?
GPT-6 Astra: 1,050,000 tokens, against 131,072 for gpt-oss-120b.
Run GPT-6 Astra and gpt-oss-120b on your own prompt.
Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.