Model benchmarks
strata_iq3s_256k local LLM performance
As of November 2026, strata_iq3s_256k runs at up to 40.2 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
40.2 tok/s
Average speed
40.0 tok/s
Avg PP
387.5 tok/s
Min memory
97.8 GB
Max context
65,536 tokens
Avg output / run
57,715 tokens
Avg runtime / run
23m 9s
Avg quality
86.4
Benchmark runs
2
GPUs tested
1
Quality by task
Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 86.1 | 86.4 | 86.8 |
| Agent Workflow | 91.5 | 91.5 | 91.5 |
| Code Generation | 72.6 | 74.2 | 75.7 |
| Role Play & Narrative | 90.2 | 91.5 | 92.9 |
| Research & Analysis | 87.4 | 88.6 | 89.8 |
Performance by hardware and tool
Every hardware/tool/quantization combination strata_iq3s_256k has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon RX 7900 XTX | OpenAI-compatible | — | 40.2 tok/s | 40.0 tok/s | 97.8 GB | 65,536 tokens | 86.4 | 2 |
Benchmark runs
All 2 strata_iq3s_256k runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is strata_iq3s_256k good for coding?
- In our benchmarks, strata_iq3s_256k scores 74.2/100 for coding. It runs at about 40.0 tok/s, so if you want more speed, local-qwen is faster (~60.2 tok/s) and still scores well for coding (84.8/100).
- Is strata_iq3s_256k good for agentic (tool-using) tasks?
- In our benchmarks, strata_iq3s_256k scores 91.5/100 for agentic workflows. It runs at about 40.0 tok/s, so if you want more speed, granite-4.2-8b-Q4_K_L is faster (~71.6 tok/s) and still scores well for agentic workflows (92.9/100).
- How fast is strata_iq3s_256k for local inference?
- Across 2 community benchmark runs, strata_iq3s_256k reaches up to 40.2 tok/s and averages 40.0 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
- How much memory does strata_iq3s_256k need?
- The leanest observed configuration used about 97.8 GB of memory.
- Which tools have been used to run strata_iq3s_256k?
- Benchmarks were submitted using OpenAI-compatible. Results are community-contributed and updated as new runs arrive.