C4ai Command A 03 2025 — Benchmarks

Benchmark scores for C4ai Command A 03 2025 aggregated from public leaderboards (4 boards), with how it ranks among open models and how it compares with its nearest neighbours. See hardware requirements for what you need to run it.

llmrun Composite Score →

79.8

90% range 72.8–86.9 · from 3 boards

CodingNot enough data yet
AgentsNot enough data yet
MathNot enough data yet
ScienceNot enough data yet
Reasoning83.490% range 74–93

C4ai Command A 03 2025 reasoning benchmarks

Multi-step logic and problem-solving — can the model think through a hard problem rather than recall an answer.

DTBench
DTBench is a set of handcrafted, expert-validated multiple-choice questions on decision theory — Newcomb-like problems and related puzzles — built by the Conceptual Reasoning Index project with Anthropic. It tests whether a model can follow an abstract argument rather than recall a fact.
61.3%
61.3%
median 67.7%open leader 93.9%
Overall best: Claude Opus 5.5 Max 98.9%
37th percentile · #43 of 68 open · #166 of 214 all
#43 of 68
#166 of 214
Lab-reported
Epoch AI · Mar 2025
LMCA
LMCA (Language Model Conceptual Argumentation) scores models on expert-rated arguments across philosophy, decision theory, and AI risk — domains where there is no empirical feedback loop to check an answer against. It is part of the Conceptual Reasoning Index, built with Anthropic by researchers Emery Cooper and Caspar Oesterheld.
10.3%
10.3%
median 24.7%open leader 55.5%
Overall best: Claude Opus 5.5 Max 68.2%
13th percentile · #47 of 54 open · #162 of 176 all
#47 of 54
#162 of 176
Lab-reported
Epoch AI · Mar 2025

C4ai Command A 03 2025 coding benchmarks

Writing, editing, and fixing real code — measured by whether the resulting program actually runs and passes tests.

Aider Polyglot
The Aider Polyglot benchmark measures real-world coding across several programming languages: the model edits code to solve Exercism exercises, and is scored on whether the final solution actually runs and passes the tests.
12.0%
12.0%
median 45.1%open leader 74.2%
Overall best: GPT 5 (Aug 07, 2025, high) 88.0%
12th percentile · #16 of 18 open · #64 of 69 all
#16 of 18
#64 of 69
Independent
Epoch AI · Mar 2025

C4ai Command A 03 2025 instruction following benchmarks

How precisely a model obeys explicit, verifiable instructions — format, length, and content constraints — and, more broadly, how much people prefer its responses head-to-head in blind human voting.

LMArena Text
Arena (arena.ai, formerly LMArena / Chatbot Arena) ranks models by millions of blind, head-to-head human votes on real prompts. The Arena Score shown here is a Bradley-Terry rating computed with Arena's default style control, so higher means people preferred that model's answers more often — it's a relative rating, not a percentage, so compare it only to other models on this board.
1353.7
1353.7
median 1321.5open leader 1487.1
Overall best: Gemini 4 Argon (High) 1523.9
61st percentile · #73 of 187 open · #203 of 405 all
#73 of 187
#203 of 405
Arena
Arena · Oct 2026

Score vs VRAM in the ≤96 GB tier

C4ai Command A 03 2025
Other models
Best Score for its VRAM
x: VRAM (GB, log scale) · y: llmrun Score

#44 of 60 that fit ≤96 GB

Each dot is a model at the best quant that fits ≤96 GB. The line joins the models with the best Score for their VRAM; a dot below it is beaten by a model that needs no more memory.

Full ≤96 GB ranking →

C4ai Command A 03 2025 vs its neighbours

Same boards for the nearest open models per llmrun Score. The highest score in each row is marked.

Board	C4ai Command A 03 2025this model	Qwen3.8 27Bbest in classQ4_K_M · 18 GB	Mistral Large Instruct 2407nearest by ScoreQ4_K_M · 86 GB	Llama 3.3 70B Instructnearest by ScoreQ4_K_M · 51 GB	Llama 3.1 70B Instructnearest by ScoreQ4_K_M · 51 GB
llmrun Score	79.8	122.1	81.1	79.7	77.9
DTBench	61.3%	88.0%	61.2%	59.5%	60.0%
LMCA	10.3%	41.4%	—	17.5%	14.8%
LMArena Text	1353.7	1437.8	1314.4	1318.1	1293.4
Compare head to head		Compare	Compare	Compare	Compare

Scores aggregated from public benchmark sources (Epoch AI: 3 boards, Arena: 1 board) — each linked from the benchmark pages. llmrun does not run these benchmarks. Percentiles and ranks are among open models in the llmrun catalogue; the median and leader marks refer to the same field.