Known-issues sheets for local language models Rev 0.1.0 · reviewed every 7 days
llm-picker

Model datasheet · DeepSeek R1 distill

DeepSeek R1-0528 Qwen3-8B

R1's reasoning distilled into a Qwen3-8B body: frontier-class competition math in 8B — AIME'24 86.0 matches models 30× its size. A specialist: brilliant at math and logic, awkward at everything agentic.

Vendor
DeepSeek
Architecture
Dense · 8B
Context
131,072 tokens
License
MIT
Released
2025-05-29
Vision
No

→ GGUF (llama.cpp / LM Studio / Ollama) → MLX (Apple Silicon)

§1 Characteristics by quantization band

Q2–Q3 — 2–3 bit

4GB weights · e.g. Q3_K_M + KV: 1.1GB at 8K · 4.5GB at 32K · 18GB at 128K

— Not yet rated · contributions welcome —

Q4–Q5 — 4–5 bit

5GB weights · e.g. Q4_K_M , mlx-4bit + KV: 1.1GB at 8K · 4.5GB at 32K · 18GB at 128K

— Not yet rated · contributions welcome —

Q6–Q8 — 6–8 bit

7GB weights · e.g. Q6_K , Q8_0 + KV: 1.1GB at 8K · 4.5GB at 32K · 18GB at 128K

— Not yet rated · contributions welcome —

FP16 — full precision (fp16/bf16)

16GB weights · e.g. bf16 + KV: 1.1GB at 8K · 4.5GB at 32K · 18GB at 128K
Math & reasoning 8/10
data checked Aug 2026
  • vendor-model-card — AIME 2024 = 86 @ bf16 · vendor-reported · 2026-08-30

    AIME 2025: 76.3, HMMT Feb 2025: 61.5 — vs base Qwen3-8B's 76.0 AIME'24. GPQA-Diamond (61.1) actually drops slightly below base.

Coding 6/10
data checked Aug 2026
  • vendor-model-card — LiveCodeBench 2408–2505 = 60.5 @ bf16 · vendor-reported · 2026-08-30

    Benchmark-decent, but community reports real-world coding friction (see known issues). Absent from Aider (only full R1 appears at 71.4%).

§2 Known issues & what fixes them

ISSUE-01 severe Q2–Q3 / Q4–Q5 / Q6–Q8 / FP16 · tool-calling

Unreliable tool/function calling — broken or ignored tool schemas are a recurring report across runtimes. This is not an agent model.

Fix
Update your runtime [weak]

Community-fixed chat templates (unsloth's GGUFs) improve it somewhat.

No fix
No known fix — choose a different model

For agents, use base Qwen3-8B and keep this one for pure reasoning.

Evidence · 2 sources · community-consensus
ISSUE-02 moderate Q2–Q3 / Q4–Q5 / Q6–Q8 / FP16 · instruction-following

Very long, redundant chains of thought — it can spend thousands of tokens re-deriving a simple answer, which is slow and floods small context windows.

Fix
Prompt technique [partial]

temp 0.6 per the card; cap max output tokens; save it for problems that deserve the thinking.

Evidence · 2 sources · community-consensus