Head-to-head comparison across 1benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
MiniMax M2.7
52
Ornith-1.0-35B
67
Pick Ornith-1.0-35B if you want the stronger benchmark profile. MiniMax M2.7 only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.
Agentic
+7.2 difference
MiniMax M2.7
Ornith-1.0-35B
$0.3 / $1.2
$0 / $0
45 t/s
N/A
2.53s
N/A
200K
262K
Pick Ornith-1.0-35B if you want the stronger benchmark profile. MiniMax M2.7 only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.
Ornith-1.0-35B is clearly ahead on the provisional aggregate, 67 to 52. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Ornith-1.0-35B's sharpest advantage is in agentic, where it averages 64.2 against 57. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 57% to 64.2%.
MiniMax M2.7 is also the more expensive model on tokens at $0.30 input / $1.20 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Ornith-1.0-35B. That is roughly Infinityx on output cost alone. Ornith-1.0-35B is the reasoning model in the pair, while MiniMax M2.7 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Ornith-1.0-35B gives you the larger context window at 262K, compared with 200K for MiniMax M2.7.
Ornith-1.0-35B is ahead on BenchLM's provisional leaderboard, 67 to 52. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 57% and 64.2%.
Ornith-1.0-35B has the edge for agentic tasks in this comparison, averaging 64.2 versus 57. Inside this category, Claw-Eval is the benchmark that creates the most daylight between them.
For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.