Claude Sonnet 5.5 vs MiniMax M3.1 Flash
Different efforts: high · thinking
Terminal-Bench 4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Sonnet 5.543.9
MiniMax M3.1 Flash25.8
FrontierCode 1.1 FrontierCode 1.1: Agents implement changes in real repositories, graded against maintainer-defined requirements
Sonnet 5.544.3
MiniMax M3.1 Flash38.6
WebDev Arena WebDev Arena: Human pairwise preferences in the stated webdev vote pool
Sonnet 5.51,710
MiniMax M3.1 Flash1,587
SciCode SciCode: Writing code for real scientific computing problems (physics, biology, math)
Sonnet 5.553.7
MiniMax M3.1 Flash53.4
DeepSWE 1.1 DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition
Sonnet 5.5 · max70.5
MiniMax M3.1 Flash—
SWE-bench Pro (v2 full) SWE-bench Pro (v2 full): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Sonnet 5.5 · xhigh95.8
MiniMax M3.1 Flash—
SWE-bench Pro (v2 hard) SWE-bench Pro (v2 hard): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Sonnet 5.5 · xhigh70.6
MiniMax M3.1 Flash—
Terminal-Bench Science 0.1 Terminal-Bench Science 0.1: Agents complete expert-authored research workflows across five scientific domains in a terminal
Sonnet 5.534.3
MiniMax M3.1 Flash—
| Benchmark | Sonnet 5.5 | MiniMax M3.1 Flash |
|---|---|---|
| Terminal-Bench 4.0 | 43.9 | 25.8 |
| FrontierCode 1.1 | 44.3 | 38.6 |
| WebDev Arena | 1,710 | 1,587 |
| SciCode | 53.7 | 53.4 |
| DeepSWE 1.1 | 70.5 · max | — |
| SWE-bench Pro (v2 full) | 95.8 · xhigh | — |
| SWE-bench Pro (v2 hard) | 70.6 · xhigh | — |
| Terminal-Bench Science 0.1 | 34.3 | — |
Specs
| Spec | Sonnet 5.5 | MiniMax M3.1 Flash |
|---|---|---|
| Input, per 1M | $2 | $0.15 |
| Output, per 1M | $10 | $0.60 |
| Cost per task? | $0.88 | $0.22 |
| Speed, tok/s | 106 | 105 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | Sep 2026 | Oct 2026 |
| See images | ✓ | ✓ |
| Read PDFs | ✓ | — |
| Hear audio | — | — |
| Watch video | — | ✓ |
| Call tools | ✓ | Partial |
| Force a tool call | — | ? |
| Return strict JSON | ✓ | Partial |
| Search the web | ✓ | ? |
| Run code | ✓ | ? |
| Use a computer | ✓ | ? |
What breaks when you switch
Sonnet 5.5 → MiniMax M3.1 Flash
No PDF input
Complete model response must be appended›
In multi-turn tools, preserve the complete assistant response, including reasoning_content and tool_calls, to maintain reasoning continuity.
MiniMax M3.1 Flash → Sonnet 5.5
No forced tool use›
Forced tool_choice any or tool returns HTTP 400; remove the forced choice.
Replace disabled thinking›
Disabled thinking returns HTTP 400; use between_tools for the lowest thinking setting.
No manual thinking budgets›
Manual thinking budgets and extra between_tools fields return HTTP 400; remove manual budgets and send between_tools without display, budget_tokens or block_binding.
Remove sampling parameters›
Non-default temperature, top_p or top_k returns HTTP 400; remove these parameters.
No assistant prefill›
A prefilled final assistant turn returns HTTP 400; remove that prefill.
Old computer tool rejected · Claude API / Google Cloud›
The computer_20251124 tool returns HTTP 400 on the Claude API and Google Cloud; use computer_toolset_20260801 there.