Compare / head-to-head
Jamba2 Minivs
Qwen3.5 9B
Jamba2 Mini and Qwen3.5 9B split their 2 shared benchmarks evenly. Qwen3.5 9B has the larger context window (262K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
1 – 1
even split
Cheaper per token
—
list price, input + output
Larger context
Qwen3.5 9B
262K tokens
Providers
2 providers
AI21 Labs · Qwen
Jamba2 Mini
AI21 Labs · released 2026-01-08
text
Qwen3.5 9B
Qwen · released 2026-03-02
textvisionvideo
- Context
- 262K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 18 · 5 core
Quality
Benchmark matrix
2 shared · 2 only Jamba2 Mini · 15 only Qwen3.5 9B| Benchmark | Jamba2 Mini | Qwen3.5 9B | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 2 | ||||
| IFBench | 38% ↗ | 64.5% ↗ | -26.5 pt | Qwen3.5 9B |
| IFEval | 97% ↗ | 91.5% ↗ | +5.5 pt | Jamba2 Mini |
| Only Jamba2 Mini reports · 2 | ||||
| Collie | 33% ↗ | not reported | — | — |
| FACTS | 57% ↗ | not reported | — | — |
| Only Qwen3.5 9B reports · 15 | ||||
| GPQA Diamond | not reported | 81.7% ↗ | — | — |
| MMLU-Pro | not reported | 82.5% ↗ | — | — |
| BFCL-V4 | not reported | 66.1% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 12% ↗epoch run | — | — |
| HMMT Feb 25 | not reported | 83.2% ↗ | — | — |
| LiveCodeBench v6 | not reported | 65.6% ↗ | — | — |
| LongBench v2 | not reported | 55.2% ↗ | — | — |
| MMLU-Redux | not reported | 91.1% ↗ | — | — |
| MMMLU | not reported | 81.2% ↗ | — | — |
| MMMU | not reported | 78.4% ↗ | — | — |
| MMMU-Pro | not reported | 70.1% ↗ | — | — |
| OSWorld-Verified | not reported | 41.8% ↗ | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | not reported | 61.7% ↗epoch run | — | — |
| SuperGPQA | not reported | 58.2% ↗ | — | — |
| TAU2-Bench | not reported | 79.1% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Jamba2 Mini minus Qwen3.5 9B in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Jamba2 Mini | Qwen3.5 9B | Edge |
|---|---|---|---|
| Context window | 256K tokens | 262K tokens | Qwen3.5 9B |
| Max output | 4K tokens | — | — |
| Input price / 1M | $0.2 ↗ | — | — |
| Output price / 1M | $0.4 ↗ | — | — |
| Cached input / 1M | — | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $0.6 | — | — |
| Modalities | text | text · vision · video | Qwen3.5 9B |
| Released | 2026-01-08 | 2026-03-02 | — |
| Cited benchmark scores | 4 | 18 | — |
Reliability
Provider status
Live from /statusMore matchups
Jamba2 Mini vs …
Models sharing the most benchmarksMore matchups
Qwen3.5 9B vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Jamba2 Mini and Qwen3.5 9B on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.