diff --git a/backend/app/autotune.py b/backend/app/autotune.py
new file mode 100644
index 0000000..68dcfbf
--- /dev/null
+++ b/backend/app/autotune.py
@@ -0,0 +1,235 @@
+"""Auto-réglage : exploite au mieux chaque modèle selon le GPU disponible.
+
+Principe :
+  1. on détecte la VRAM du GPU (nvidia-smi, sinon rocm-smi, sinon override env) ;
+  2. on lit les métadonnées du modèle via Ollama (/api/show + /api/tags) :
+     contexte max, architecture (couches, têtes KV, dimension), taille sur disque ;
+  3. on calcule la fenêtre de contexte (num_ctx) la plus grande qui tient en VRAM,
+     via une estimation du cache KV, puis un nombre de jetons de sortie cohérent.
+
+Tout est best-effort : si une info manque, on retombe sur des paliers prudents.
+"""
+from __future__ import annotations
+
+import shutil
+import subprocess
+
+import httpx
+
+from .config import settings
+from .ollama_client import ollama
+
+# Paliers de contexte « ronds » proposés (bornés par le contexte du modèle).
+CTX_STEPS = [2048, 4096, 8192, 12288, 16384, 24576, 32768, 49152, 65536, 131072]
+
+# Marges VRAM (Mo) : OS/driver + buffers de calcul d'Ollama.
+VRAM_OVERHEAD_MB = 1024
+VRAM_COMPUTE_BUFFER_MB = 768
+
+
+# ── Détection GPU ────────────────────────────────────────────────────────
+def detect_gpu() -> dict:
+    """Renvoie {available, name, vram_total_mb, source}."""
+    # 1) Override explicite (env) — prioritaire, utile en conteneur sans GPU.
+    if settings.gpu_vram_mb > 0:
+        return {
+            "available": True,
+            "name": settings.gpu_name or "GPU (configuré)",
+            "vram_total_mb": settings.gpu_vram_mb,
+            "source": "env",
+        }
+
+    # 2) NVIDIA
+    if shutil.which("nvidia-smi"):
+        try:
+            out = subprocess.run(
+                ["nvidia-smi",
+                 "--query-gpu=name,memory.total",
+                 "--format=csv,noheader,nounits"],
+                capture_output=True, text=True, timeout=5,
+            )
+            if out.returncode == 0 and out.stdout.strip():
+                line = out.stdout.strip().splitlines()[0]
+                name, total = [p.strip() for p in line.split(",")]
+                return {
+                    "available": True,
+                    "name": name,
+                    "vram_total_mb": int(float(total)),
+                    "source": "nvidia-smi",
+                }
+        except (OSError, ValueError, subprocess.SubprocessError):
+            pass
+
+    # 3) AMD (ROCm) — best-effort
+    if shutil.which("rocm-smi"):
+        try:
+            out = subprocess.run(
+                ["rocm-smi", "--showmeminfo", "vram", "--csv"],
+                capture_output=True, text=True, timeout=5,
+            )
+            if out.returncode == 0:
+                # Cherche le plus grand entier (octets) -> Mo
+                nums = [int(x) for x in out.stdout.replace(",", " ").split()
+                        if x.isdigit()]
+                if nums:
+                    return {
+                        "available": True,
+                        "name": "GPU AMD",
+                        "vram_total_mb": max(nums) // (1024 * 1024),
+                        "source": "rocm-smi",
+                    }
+        except (OSError, ValueError, subprocess.SubprocessError):
+            pass
+
+    return {"available": False, "name": "CPU", "vram_total_mb": 0, "source": "none"}
+
+
+# ── Métadonnées modèle ───────────────────────────────────────────────────
+async def model_profile(model: str) -> dict:
+    """Extrait contexte max, archi (couches/têtes/dim), taille disque (Mo)."""
+    profile = {
+        "context_length": None,
+        "block_count": None,
+        "head_count": None,
+        "head_count_kv": None,
+        "embedding_length": None,
+        "size_mb": None,
+        "parameter_size": None,
+        "quantization": None,
+    }
+    try:
+        show = await ollama.show(model)
+    except (httpx.HTTPError, OSError):
+        return profile
+
+    info = show.get("model_info", {}) or {}
+    arch = info.get("general.architecture", "")
+
+    def pick(*suffixes):
+        for s in suffixes:
+            key = f"{arch}.{s}" if arch else s
+            if key in info:
+                return info[key]
+            for k, v in info.items():
+                if k.endswith(s):
+                    return v
+        return None
+
+    profile["context_length"] = pick("context_length")
+    profile["block_count"] = pick("block_count")
+    profile["head_count"] = pick("attention.head_count")
+    profile["head_count_kv"] = pick("attention.head_count_kv")
+    profile["embedding_length"] = pick("embedding_length")
+
+    details = show.get("details", {}) or {}
+    profile["parameter_size"] = details.get("parameter_size")
+    profile["quantization"] = details.get("quantization_level")
+
+    # Taille sur disque via /api/tags
+    try:
+        for m in await ollama.list_models():
+            if m.get("name") == model:
+                profile["size_mb"] = int(m.get("size", 0)) // (1024 * 1024)
+                break
+    except (httpx.HTTPError, OSError):
+        pass
+
+    return profile
+
+
+def _kv_bytes_per_token(p: dict) -> int | None:
+    """Estimation des octets de cache KV par jeton (KV en f16)."""
+    blocks = p.get("block_count")
+    kv_heads = p.get("head_count_kv")
+    heads = p.get("head_count")
+    emb = p.get("embedding_length")
+    if not all((blocks, kv_heads, heads, emb)):
+        return None
+    head_dim = emb / heads
+    # 2 (clé+valeur) * couches * têtes_kv * dim_tête * 2 octets (f16)
+    return int(2 * blocks * kv_heads * head_dim * 2)
+
+
+def _round_ctx(candidate: int, ctx_max: int | None) -> int:
+    cap = ctx_max or CTX_STEPS[-1]
+    best = CTX_STEPS[0]
+    for step in CTX_STEPS:
+        if step <= candidate and step <= cap:
+            best = step
+    # Si le modèle plafonne bas, respecte son contexte max.
+    return min(best, cap)
+
+
+# ── Recommandation ───────────────────────────────────────────────────────
+async def recommend(model: str) -> dict:
+    gpu = detect_gpu()
+    prof = await model_profile(model)
+    ctx_max = prof.get("context_length")
+
+    rationale: list[str] = []
+
+    if not gpu["available"]:
+        # CPU : on reste prudent (le contexte coûte cher en latence).
+        num_ctx = _round_ctx(8192, ctx_max)
+        rationale.append("Aucun GPU détecté — réglages CPU prudents.")
+    else:
+        vram = gpu["vram_total_mb"]
+        model_mb = prof.get("size_mb") or _fallback_model_mb(prof)
+        budget = vram - model_mb - VRAM_OVERHEAD_MB - VRAM_COMPUTE_BUFFER_MB
+        kv_per_tok = _kv_bytes_per_token(prof)
+
+        if budget <= 256:
+            num_ctx = _round_ctx(2048, ctx_max)
+            rationale.append("VRAM insuffisante pour les poids — contexte minimal.")
+        elif kv_per_tok:
+            tokens_fit = int(budget * 1024 * 1024 / kv_per_tok)
+            num_ctx = _round_ctx(tokens_fit, ctx_max)
+            rationale.append(
+                f"{vram} Mo VRAM − {model_mb} Mo poids → budget KV "
+                f"{budget} Mo (~{kv_per_tok // 1024} Ko/jeton)."
+            )
+        else:
+            # Archi inconnue : paliers selon le budget VRAM restant.
+            num_ctx = _round_ctx(_tier_ctx(budget), ctx_max)
+            rationale.append("Archi modèle incomplète — estimation par paliers.")
+
+    # Jetons de sortie : moitié du contexte, borné, en laissant de la place au prompt.
+    max_tokens = max(512, min(num_ctx // 2, 8192))
+
+    return {
+        "gpu": gpu,
+        "model": model,
+        "model_profile": {
+            "context_length": ctx_max,
+            "parameter_size": prof.get("parameter_size"),
+            "quantization": prof.get("quantization"),
+            "size_mb": prof.get("size_mb"),
+        },
+        "recommended": {"num_ctx": num_ctx, "max_tokens": max_tokens},
+        "rationale": " ".join(rationale),
+    }
+
+
+def _fallback_model_mb(prof: dict) -> int:
+    """Estime la taille des poids si /api/tags n'a rien donné."""
+    ps = (prof.get("parameter_size") or "").upper().replace("B", "")
+    try:
+        billions = float(ps)
+    except ValueError:
+        billions = 8.0
+    # ~0.6 Go/milliard en Q4, approximation prudente.
+    return int(billions * 600)
+
+
+def _tier_ctx(budget_mb: int) -> int:
+    if budget_mb >= 12000:
+        return 32768
+    if budget_mb >= 8000:
+        return 16384
+    if budget_mb >= 5000:
+        return 12288
+    if budget_mb >= 3000:
+        return 8192
+    if budget_mb >= 1500:
+        return 4096
+    return 2048
diff --git a/backend/app/config.py b/backend/app/config.py
index 5ccc324..d3fb6fa 100644
--- a/backend/app/config.py
+++ b/backend/app/config.py
@@ -11,6 +11,11 @@ class Settings(BaseSettings):
     data_dir: str = "/data"
     port: int = 8080
 
+    # Override manuel de la VRAM (Mo) si la détection GPU échoue dans le
+    # conteneur (utile sur Unraid où le GPU est sur l'hôte / un autre conteneur).
+    gpu_vram_mb: int = 0
+    gpu_name: str = ""
+
     model_config = SettingsConfigDict(env_file=".env", extra="ignore")
 
 
diff --git a/backend/app/ollama_client.py b/backend/app/ollama_client.py
index 91bb492..50426c9 100644
--- a/backend/app/ollama_client.py
+++ b/backend/app/ollama_client.py
@@ -33,6 +33,13 @@ class OllamaClient:
             resp.raise_for_status()
             return resp.json().get("models", [])
 
+    async def show(self, name: str) -> dict:
+        """Métadonnées détaillées d'un modèle (/api/show)."""
+        async with httpx.AsyncClient(timeout=15.0) as client:
+            resp = await client.post(f"{self.host}/api/show", json={"name": name})
+            resp.raise_for_status()
+            return resp.json()
+
     async def pull_model(self, name: str) -> AsyncIterator[dict]:
         """Télécharge un modèle en streamant la progression (/api/pull)."""
         async with httpx.AsyncClient(timeout=None) as client: