INTELLIGENCE
12.0
Media Cap quality axis
Alibaba · Open weightsQwen3 235B A22b Instruct 2507 Fp8
Active
Model portfolio
Qwen3 235B A22b Instruct 2507 Fp8
Tap a star to rate
Alibaba · Open · Context 262K tokens
Language Cap
45
Cadle Language Cap 45 · 82%
Benchmark average
56.9
Not enough benchmark results yet
Pricing
$0.20
$0.60 / 1M out · 60.5 pts/$
CADLE CAP vs MODEL MATRIX
Cadle Cap ranks models on every published score we have for them — harder benchmarks weigh more. Missing tests are skipped, not zeroed. It answers who is stronger than whom in today's pool.
Model Matrix is the datasheet: Overall, Intelligence, Coding, Agentic, MMLU, GPQA, Math, LiveCodeBench, HumanEval, Reason, ARC-AGI. A dash means unmeasured. AVG is an importance-weighted mean of that model's filled cells only.
MODEL MATRIX · TRUTH CENTER CELLS
Overall
No data yet
Intelligence
12.0
Coding
4.7
Agentic
42.7
MMLU
82.8
GPQA
75.3
Math
98.0
LiveCodeBench
52.4
HumanEval
No data yet
Reason
46.0
ARC-AGI
No data yet
Coverage AVG
56.9
Cadle Cap ranks models with within-benchmark percentiles so ARC-AGI ~40 and MMLU ~90 never compete raw. Coverage AVG is an importance-weighted mean of filled expert cells only — blanks are skipped, not zeroed.
ANALYSIS · TRUTH CENTER
Hard benches (GPQA · ARC-AGI · LCB · Agentic) sit left — a mid-40 on ARC-AGI can still be frontier. Broader benches on the right are easier and often higher.
CADLE INSIGHTS
Only measured signals
CODING
4.7
Media Cap quality axis
AGENTIC
42.7
Media Cap quality axis
MATH
98.0
Media Cap quality axis
STRONGEST SIGNAL
Intelligence
Cap axis 12.0
BLENDED $/1M
$0.80
In $0.20 · Out $0.60
VALUE INDEX
71.1
Quality ÷ blended $/1M (higher is better)
CONTEXT
262K
Provider context window (not max output)
SPEED · COST · QUALITY
Measured matrix / AA / gateway rows only.
- Cap Intelligence
- 48.4
- AA / Intel index
- 12.0
- Cap Coding
- 1.2
- Matrix Coding
- 4.7
- Input $/1M
- $0.20
- Output $/1M
- $0.60
- Value pts/$
- 60.55
DESCRIPTION + DETAILS
PROVIDER
Alibaba
CONTEXT
262K tokens
STATUS
ACTIVE
COMMUNITY REVIEWS
OPERATOR NOTES
Quality, latency, cost, and failure modes — not marketing copy.
No reviews yet
WRITE A REVIEW
Tap a star to rate
Stored on this device · synced when signed in
No operator notes yet
First review sets the tone for how teams evaluate Qwen3 235B A22b Instruct 2507 Fp8.
MODEL COMPARISON
| Metric | |
|---|---|
| Provider | Alibaba |
| Release | |
| Input / 1M | $0.20 |
| Output / 1M | $0.60 |
| Context | 262K tokens |
| Cadle Cap | 45 |
| Cap board | Cadle Language Cap 45 |
DATA PROVENANCE
Scores may use different scales — check the Benchmarks tab for source links.
- Supabase models
- Cadle public observations
SOURCE TRACKING
Evidence linked to this model — listing, hub, benchmarks, or live gateway prices.
API
Alibaba listing
MEASURED CELLS
Same Truth Center numbers as the datasheet above. Accuracy, reliability, safety, and cost scores are not invented.
AA IQ12
Coding4.7
Agentic42.7
MMLU82.8
GPQA75.3
Math98
LiveCodeBench52.4