For coding agents, GPT-5.6 Sol costs 14x as much as Codestral per request ($0.099 vs $0.0070 on our measured traffic), because GPT-5.6 Sol lists at $4 input, $0.40 cached input and $20 output per million tokens against $0.30, $0.03 and $0.90 for Codestral, and GPT-5.6 Sol charges more for prompts over 272K tokens.
Run GPT-5.6 Sol in your agent from $19/month
See plans| Cost at list price | Codestral | GPT-5.6 Sol |
|---|---|---|
| Prompt under 100K tokens50% of requests, 87% of input cached | $0.0043 | $0.062 |
| Prompt of 100K to 272K39% of requests, 88% of input cached | $0.011 | $0.15 |
| Prompt over 272K tokens11% of requests, 86% of input cached | n/a | n/a |
| Average request | $0.0070 | $0.099 |
| 100 requests | $0.70 | $9.94 |
| Median session16 requests; a session ends after 10 idle minutes | $0.084 | $1.19 |
| A month of regular usemedian of 20 users active on 5+ days | $8.91 | $125 |
Each logged request priced at the vendor’s list rates for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request. Both models are priced on the requests up to 256K tokens that fit in both context windows; band shares above describe all requests.
Codestral's 256K window cannot take 13% of coding-agent requests, so the comparison uses only requests both can take. GPT-5.6 Sol is superseded: GPT-6.1 Sol costs $0.067 per request.
Pick a winner in each dimension. You can change your vote any time. 1 vote cast so far.
Correctness and depth of what it produces
Tool calls, instruction-following, and multi-step tasks
Tokens per second and time-to-first-token
How much capability you get per dollar
Consistent results: fewer refusals, loops and format breaks
Setting it up? GPT-5.6 Sol: model ID, providers and agent config
The agent matters as much as the model. Compare AI agents →
Current Claude, GPT, Gemini and open models are included with Standard Compute