Model benchmarks
HermesFableAgentThink local LLM performance
As of September 2026, HermesFableAgentThink runs at up to 21.7 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
21.7 tok/s
Average speed
21.3 tok/s
Avg PP
355.6 tok/s
Min memory
33.3 GB
Max context
262.144 tokens
Avg output / run
14.242 tokens
Avg runtime / run
12m 3s
Avg quality
82.1
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 | 78.6 | 82.1 | 85.6 |
| Agent Workflow | 76.4 | 82.1 | 87.8 |
| Code Generation | 70.1 | 72.4 | 74.7 |
| Role Play & Narrative | 81.2 | 87.8 | 94.4 |
| Research & Analysis | 85.4 | 86.0 | 86.7 |
Performance by hardware and tool
Every hardware/tool/quantization combination HermesFableAgentThink has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M5 Max | oMLX | — | 21.7 tok/s | 21.3 tok/s | 33.3 GB | 262.144 tokens | 82.1 | 2 |
Benchmark runs
All 2 HermesFableAgentThink runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is HermesFableAgentThink good for coding?
- In our benchmarks, HermesFableAgentThink scores 72.4/100 for coding. It runs at about 21.3 tok/s, so if you want more speed, /home/polaris/models/Tiel-Coder-35B-A3B-MTP-UD-Q6_K_XL.gguf is faster (~35.6 tok/s) and still scores well for coding (84.9/100).
- Is HermesFableAgentThink good for agentic (tool-using) tasks?
- In our benchmarks, HermesFableAgentThink scores 82.1/100 for agentic workflows. It runs at about 21.3 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 HermesFableAgentThink for local inference?
- Across 2 community benchmark runs, HermesFableAgentThink reaches up to 21.7 tok/s and averages 21.3 tok/s, with the fastest results on Apple M5 Max.
- How much memory does HermesFableAgentThink need?
- The leanest observed configuration used about 33.3 GB of memory.
- Which tools have been used to run HermesFableAgentThink?
- Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.