GPT-6 Astra vs Mistral Medium 3.5
API pricing, context window and capabilities for GPT-6 Astra and Mistral Medium 3.5, side by side. Prices are the cheapest listed serving route per 1M tokens, from LiteLLM and OpenRouter.
Verdict
computed from the published numbersGPT-6 Astra wins on Terminal-Bench, on intelligence and on context window. Mistral Medium 3.5 wins on price.
Price
Mistral Medium 3.5
Terminal-Bench
GPT-6 Astra
Intelligence
GPT-6 Astra
Context window
GPT-6 Astra
Key takeaways
who wins on what, from the published numbersGPT-6 Astra wins
- Larger context window (1.1M vs 262K)
- More serving providers (5 vs 4)
- Supports prompt caching
- Has web search
Mistral Medium 3.5 wins
- Cheaper input ($10.00 vs $1.50 /1M)
- Cheaper output ($50.00 vs $7.50 /1M)
- Higher max output (128K vs 262K)
Cheaper overall
Mistral Medium 3.5
$1.50 + $7.50 /1M
Larger context
GPT-6 Astra
1.1M
More providers
GPT-6 Astra
5 providers
Pricing & specs
cheapest listed route · per 1M tokensOn-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments can be materially lower.
| GPT-6 Astra | Mistral Medium 3.5 | |
|---|---|---|
| Input $/1M | $10.00via OpenAI | $1.50via Mistral best |
| Output $/1M | $50.00via OpenAI | $7.50via Mistral best |
| Context window | 1.1M best | 262K |
| Max output | 128K | 262K best |
| Serving providers | 5 best | 4 |
| Date | Released Sep 4, 2026 | Released Apr 30, 2026 |
Benchmarks
Benchmark scores, side by side| GPT-6 Astra | Mistral Medium 3.5 | |
|---|---|---|
| Intelligence | 52.7 best | 14.2 |
| Terminal-Bench | 88% best | 51% |
| GPQA Diamond | 96% best | 75% |
| Output speed | 61 t/s | 167 t/s best |
| Latency (TTFT) | 280.05s | 2.21s best |
Cost calculator
choose a serving route for each sideInput tokens / month
Output tokens / month
GPT-6 Astra
via OpenAI
$100
Mistral Medium 3.5
via OpenRouter
$15
Capabilities
feature support, side by side| Feature | GPT-6 Astra | Mistral Medium 3.5 |
|---|---|---|
| Vision | ✓ | ✓ |
| Tool use | ✓ | ✓ |
| Reasoning | ✓ | ✓ |
| Audio | — | — |
| Prompt caching | ✓ | — |
| Structured output | ✓ | ✓ |
| Web search | ✓ | — |
Related comparisons
Frequently asked questions
Which is cheaper, GPT-6 Astra or Mistral Medium 3.5?
Mistral Medium 3.5 is cheaper for input ($10.00 vs $1.50 per 1M tokens) from Mistral. Mistral Medium 3.5 is cheaper for output ($50.00 vs $7.50 per 1M tokens) from Mistral.
What is the context window difference between GPT-6 Astra and Mistral Medium 3.5?
GPT-6 Astra has a 1.1M-token context window and Mistral Medium 3.5 has 262K tokens, so GPT-6 Astra can take a larger prompt.
How do GPT-6 Astra and Mistral Medium 3.5 compare on benchmarks?
On the tracked benchmarks, GPT-6 Astra leads on Terminal-Bench, and GPT-6 Astra leads on intelligence.
Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI