Added fp4 quant/dequant and dequant optimizations.

bitsandbytes/cextension.py

@@ -9,7 +9,7 @@ from bitsandbytes.cuda_setup.main import CUDASetup
 
 
 setup = CUDASetup.get_instance()
-if setup.initialized != True:
+if not setup.initialized:
     setup.run_cuda_setup()
     if 'BITSANDBYTES_NOWELCOME' not in os.environ or str(os.environ['BITSANDBYTES_NOWELCOME']) == '0':
         setup.print_log_stack()

bitsandbytes/cuda_setup/main.py

@@ -35,6 +35,9 @@ class CUDASetup:
         raise RuntimeError("Call get_instance() instead")
 
     def generate_instructions(self):
+        if getattr(self, 'error', False): return
+        print(self.error)
+        self.error = True
         if self.cuda is None:
             self.add_log_entry('CUDA SETUP: Problem: The main issue seems to be that the main CUDA library was not detected.')
             self.add_log_entry('CUDA SETUP: Solution 1): Your paths are probably not up-to-date. You can update them via: sudo ldconfig.')
@@ -84,6 +87,7 @@ class CUDASetup:
             self.has_printed = False
             self.lib = None
             self.initialized = False
+            self.error = False
 
     def run_cuda_setup(self):
         self.initialized = True

bitsandbytes/functional.py

@@ -168,7 +168,8 @@ def create_fp8_map(signed=True, exponent_bits=5, precision_bits=2, total_bits=8)
     values = []
     lst = list(itertools.product([0, 1], repeat=precision_bits))
     #for ev in evalues:
-    bias = 2**(exponent_bits-1)-1
+    bias = 2**(exponent_bits-1)+1
+    print(bias)
     for evalue in range(2**(exponent_bits)):
         for bit_pattern in lst:
             value = (1 if evalue != 0 else 0)
@@ -176,10 +177,12 @@ def create_fp8_map(signed=True, exponent_bits=5, precision_bits=2, total_bits=8)
                 value += pval*(2**-(i+1))
             if evalue == 0:
                 # subnormals
-                value = value*2**-(bias-1)
+                value = value*2**-(bias)
             else:
                 # normals
-                value = value*2**-(evalue-bias-2)
+                print(value, 1)
+                value = value*2**-(evalue-bias-1)
+                print(value, 2)
             values.append(value)
             if signed:
                 values.append(-value)
@@ -193,7 +196,7 @@ def create_fp8_map(signed=True, exponent_bits=5, precision_bits=2, total_bits=8)
             values.append(0)
     values.sort()
     code = torch.Tensor(values)
-    code /= code.max()
+    #code /= code.max()
 
     return code
 
@@ -587,7 +590,7 @@ def dequantize_blockwise(
         code = code.to(A.device)
         if blocksize not in [2048, 4096, 1024, 512, 256, 128, 64]:
             raise ValueError(f"The blockwise of {blocksize} is not supported. Supported values: [2048, 4096, 1024, 512, 256, 128, 64]")
-        is_on_gpu([A, out])
+        is_on_gpu([A, absmax, out])
         if out.dtype == torch.float32:
             lib.cdequantize_blockwise_fp32(get_ptr(code), get_ptr(A), get_ptr(absmax), get_ptr(out), ct.c_int(blocksize), ct.c_int(A.numel()))
         elif out.dtype == torch.float16:
@@ -602,6 +605,116 @@ def dequantize_blockwise(
     return out
 
 
+def quantize_fp4(A: Tensor, absmax: Tensor = None, out: Tensor = None, blocksize=64) -> Tensor:
+    """
+    Quantize tensor A in blocks of FP4 values.
+
+    Quantizes tensor A by dividing it into blocks which are independently quantized to FP4.
+
+    Parameters
+    ----------
+    A : torch.Tensor
+        The input tensor.
+    absmax : torch.Tensor
+        The absmax values.
+    out : torch.Tensor
+        The output tensor (8-bit).
+    blocksize : int
+        The blocksize used in quantization.
+
+    Returns
+    -------
+    torch.Tensor:
+        The 8-bit tensor with packed 4-bit values.
+    tuple(torch.Tensor, torch.Size, torch.dtype):
+        The quantization state to undo the quantization.
+    """
+    if A.device.type != 'cuda':
+        raise NotImplementedError(f'Device type not supported for FP4 quantization: {A.device.type}')
+
+    n = A.numel()
+    input_shape = A.shape
+
+    if absmax is None:
+        blocks = n // blocksize
+        blocks += 1 if n % blocksize > 0 else 0
+        absmax = torch.zeros((blocks,), device=A.device)
+
+    state = (absmax, input_shape, A.dtype)
+
+    if out is None:
+        out = torch.zeros(((n+1)//2,), dtype=torch.uint8, device=A.device)
+
+    assert blocksize in [4096, 2048, 1024, 512, 256, 128, 64]
+
+    prev_device = pre_call(A.device)
+    is_on_gpu([A, out, absmax])
+
+    if A.dtype == torch.float32:
+        lib.cquantize_blockwise_fp32_fp4(get_ptr(None), get_ptr(A), get_ptr(absmax), get_ptr(out), ct.c_int32(blocksize), ct.c_int(n))
+    elif A.dtype == torch.float16:
+        lib.cquantize_blockwise_fp16_fp4(get_ptr(None), get_ptr(A), get_ptr(absmax), get_ptr(out), ct.c_int32(blocksize), ct.c_int(n))
+    else:
+        raise ValueError(f"Blockwise quantization only supports 16/32-bit floats, but got {A.dtype}")
+    post_call(A.device)
+
+    return out, state
+
+
+def dequantize_fp4(A: Tensor,quant_state: Tuple[Tensor, Tensor] = None, absmax: Tensor = None, out: Tensor = None, blocksize: int = 64) -> Tensor:
+    """
+    Dequantizes FP4 blockwise quantized values.
+
+    Dequantizes the tensor A with maximum absolute values absmax in blocks of size blocksize.
+
+    Parameters
+    ----------
+    A : torch.Tensor
+        The input 8-bit tensor (packed 4-bit values).
+    quant_state : tuple(torch.Tensor, torch.Size, torch.dtype)
+        Tuple of absmax values, original tensor shape and original dtype.
+    absmax : torch.Tensor
+        The absmax values.
+    out : torch.Tensor
+        Dequantized output tensor.
+
+
+    Returns
+    -------
+    torch.Tensor:
+        Dequantized tensor.
+    """
+    if blocksize not in [2048, 4096, 1024, 512, 256, 128, 64]:
+        raise ValueError(f"The blockwise of {blocksize} is not supported. Supported values: [2048, 4096, 1024, 512, 256, 128, 64]")
+
+    if quant_state is None:
+        assert absmax is not None and out is not None
+        shape = out.shape
+        dtype = out.dtype
+    else:
+        absmax, shape, dtype = quant_state
+
+
+    if out is None:
+        out = torch.empty(shape, dtype=dtype, device=A.device)
+
+    n = out.numel()
+
+    device = pre_call(A.device)
+    is_on_gpu([A, absmax, out])
+    if out.dtype == torch.float32:
+        lib.cdequantize_blockwise_fp32_fp4(get_ptr(None), get_ptr(A), get_ptr(absmax), get_ptr(out), ct.c_int(blocksize), ct.c_int(n))
+    elif out.dtype == torch.float16:
+        lib.cdequantize_blockwise_fp16_fp4(get_ptr(None), get_ptr(A), get_ptr(absmax), get_ptr(out), ct.c_int(blocksize), ct.c_int(n))
+    else:
+        raise ValueError(f"Blockwise quantization only supports 16/32-bit floats, but got {A.dtype}")
+    post_call(A.device)
+
+    return out
+
+
+
+
 def quantize(A: Tensor, code: Tensor = None, out: Tensor = None) -> Tensor:
     if code is None:
         if "dynamic" not in name2qmap:

csrc/kernels.cu

@@ -43,6 +43,79 @@ __device__ float atomicMin(float* address, float val) {
   return __int_as_float(old);
 }
 
+__device__ float dDequantizeFP4(unsigned char val, float absmax)
+{
+  float sign = (val & 0b1000) == 8 ? -1.0f : 1.0f;
+  if((val & 0b0110) == 0)
+  {
+    // subnormal
+    if((val & 0b0001) == 0)
+      return 0.0f;
+    else
+      return sign*0.0625f*absmax;
+  }
+  else
+  {
+    // normal
+    float exponent = ((val & 0b0100) == 4 ? 2.0f : 8.0f) + ((val & 0b0010) == 2 ? 0.0f : 2.0f);
+    float fraction = (val & 0b0001) == 1 ? 1.5f : 1.0f;
+
+    return sign*exponent*fraction*absmax;
+  }
+}
+
+__device__ unsigned char dQuantizeFP4(float x)
+{
+  // FP4 with bias of 3
+  // first bit is a sign
+  // subnormals
+  // 0b000 = 0
+  // 0b001 = 0.0625
+  // 0b110 = 2
+  // 0b111 = 3
+  // 0b100 = 4
+  // 0b101 = 6
+  // 0b010 = 8
+  // 0b011 = 12
+
+  int sign = x < 0 ? 0b1000 : 0b0000;
+  x = fabsf(x);
+  if(x > 3.5f)
+  {
+    if( x > 7.0f)
+    {
+      if( x > 10.0f)
+        return 0b0011+sign;
+      else
+        return 0b0010+sign;
+    }
+    else
+    {
+      if(x > 5.0f)
+        return 0b101+sign;
+      else
+        return 0b100+sign;
+    }
+  }
+  else
+  {
+    if(x > 1.03125f)
+    {
+      if(x > 2.5f)
+        return 0b0111+sign;
+      else
+        return 0b0110+sign;
+    }
+    else
+    {
+      if(x > 0.03125f)
+        return 0b0001+sign;
+      else
+        return 0b0000+sign;
+    }
+  }
+}
+
 template <int STOCHASTIC>
 __device__ unsigned char dQuantize(float* smem_code, const float rand, float x)
 {
@@ -427,7 +500,7 @@ __global__ void kQuantize(float * code, float * __restrict__ const A, unsigned c
   }
 }
 
-template<typename T, int BLOCK_SIZE, int NUM_PER_TH, int STOCHASTIC>
+template<typename T, int BLOCK_SIZE, int NUM_PER_TH, int STOCHASTIC, int FP4>
 //__launch_bounds__(TH, 4)
 __global__ void kQuantizeBlockwise(float * code, T * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n)
 {
@@ -437,13 +510,13 @@ __global__ void kQuantizeBlockwise(float * code, T * __restrict__ const A, float
 
   T vals[NUM_PER_TH];
   float rand_vals[NUM_PER_TH];
-  unsigned char qvals[NUM_PER_TH];
+  unsigned char qvals[FP4 ? NUM_PER_TH/2 : NUM_PER_TH];
   //float local_abs_max = -FLT_MAX;
   float local_abs_max = 0.0f;
   int local_rand_idx = 0;
 
   typedef cub::BlockLoad<T, BLOCK_SIZE/NUM_PER_TH, NUM_PER_TH, cub::BLOCK_LOAD_WARP_TRANSPOSE> LoadT;
-  typedef cub::BlockStore<unsigned char, BLOCK_SIZE/NUM_PER_TH, NUM_PER_TH, cub::BLOCK_STORE_WARP_TRANSPOSE> StoreChar;
+  typedef cub::BlockStore<unsigned char, BLOCK_SIZE/NUM_PER_TH, FP4 ? NUM_PER_TH/2 : NUM_PER_TH, cub::BLOCK_STORE_WARP_TRANSPOSE> StoreChar;
   typedef cub::BlockReduce<float, BLOCK_SIZE/NUM_PER_TH> BlockReduce;
   typedef cub::BlockLoad<float, BLOCK_SIZE/NUM_PER_TH, NUM_PER_TH, cub::BLOCK_LOAD_WARP_TRANSPOSE> LoadFloat;
 
@@ -454,8 +527,9 @@ __global__ void kQuantizeBlockwise(float * code, T * __restrict__ const A, float
   __shared__ float smem_code[256];
   __shared__ float smem_absmax_value[1];
 
-  for(int i = threadIdx.x; i < 256; i+=blockDim.x)
+  if(!FP4)
-    smem_code[i] = code[i];
+    for(int i = threadIdx.x; i < 256; i+=blockDim.x)
+      smem_code[i] = code[i];
 
   for (unsigned int i = base_idx; i < n_full; i += gridDim.x*BLOCK_SIZE)
   {
@@ -495,61 +569,138 @@ __global__ void kQuantizeBlockwise(float * code, T * __restrict__ const A, float
       LoadFloat(loadf).Load(&rand[local_rand_idx], rand_vals, BLOCK_SIZE, 0);
     }
 
-    #pragma unroll NUM_PER_TH
+    if(FP4)
-    for(int j = 0; j < NUM_PER_TH; j++)
     {
-      if(!STOCHASTIC)
+      #pragma unroll NUM_PER_TH
-       qvals[j] = dQuantize<0>(smem_code, 0.0f, ((float)vals[j])*local_abs_max);
+      for(int j = 0; j < NUM_PER_TH/2; j++)
-      else
+      {
-       qvals[j] = dQuantize<1>(smem_code, rand_vals[j], ((float)vals[j])*local_abs_max);
+        unsigned char packed_fp4 = 0;
+        packed_fp4 |= dQuantizeFP4(((float)vals[2*j])*local_abs_max*12.0f) << 4;
+        packed_fp4 |= dQuantizeFP4(((float)vals[2*j+1])*local_abs_max*12.0f);
+        qvals[j] = packed_fp4;
+      }
+    }
+    else
+    {
+      #pragma unroll NUM_PER_TH
+      for(int j = 0; j < NUM_PER_TH; j++)
+      {
+          if(!STOCHASTIC)
+           qvals[j] = dQuantize<0>(smem_code, 0.0f, ((float)vals[j])*local_abs_max);
+          else
+           qvals[j] = dQuantize<1>(smem_code, rand_vals[j], ((float)vals[j])*local_abs_max);
+      }
     }
 
     __syncthreads();
-    StoreChar(storec).Store(&(out[i]), qvals, valid_items);
+    StoreChar(storec).Store(&(out[FP4 ? i/2 : i]), qvals, FP4 ? (valid_items+1)/2 : valid_items);
   }
 }
 
-template<typename T, int BLOCK_SIZE, int THREADS, int NUM_PER_TH>
+template<typename T, int TILE_SIZE, int THREADS, int NUM_PER_TH, int FP4>
-__global__ void kDequantizeBlockwise(float *code, unsigned char * A, float * absmax, T *out, const int n)
+__global__ void kDequantizeBlockwise(float *code, unsigned char * A, float * absmax, T *out, const int blocksize, const int n)
 {
 
-  const int n_full = gridDim.x * BLOCK_SIZE;
+  const int n_load = (gridDim.x * TILE_SIZE);
-  int valid_items = 0;
+  int valid_items_load = 0;
-  const int base_idx = (blockIdx.x * BLOCK_SIZE);
+  int valid_items_store = 0;
+  const int base_idx = (blockIdx.x * TILE_SIZE);
 
-  T vals[NUM_PER_TH];
+  T vals[NUM_PER_TH*(FP4 ? 2 : 1)];
   unsigned char qvals[NUM_PER_TH];
   float local_abs_max = -FLT_MAX;
 
   typedef cub::BlockLoad<unsigned char, THREADS, NUM_PER_TH, cub::BLOCK_LOAD_WARP_TRANSPOSE> LoadChar;
-  typedef cub::BlockStore<T, THREADS, NUM_PER_TH, cub::BLOCK_STORE_WARP_TRANSPOSE> StoreT;
+  typedef cub::BlockStore<T, THREADS, NUM_PER_TH*(FP4 ? 2 : 1), cub::BLOCK_STORE_WARP_TRANSPOSE> StoreT;
 
   __shared__ typename LoadChar::TempStorage loadchar;
   __shared__ typename StoreT::TempStorage storet;
-  //__shared__ float smem_code[256];
-  //float local_code[16];
 
-  //if(threadIdx.x < 256)
+  for (unsigned int i = base_idx; i < n_load; i += gridDim.x*TILE_SIZE)
-    //smem_code[threadIdx.x] = code[threadIdx.x];
-
-  for (unsigned int i = base_idx; i < n_full; i += gridDim.x*BLOCK_SIZE)
   {
-      valid_items = n - i > BLOCK_SIZE ? BLOCK_SIZE : n - i;
+      if(FP4)
-      local_abs_max = absmax[i/BLOCK_SIZE];
+      {
+        valid_items_load = (n+1)/2 - i > TILE_SIZE ? TILE_SIZE : (n+1)/2 - i;
+        valid_items_store = n - i*2 > TILE_SIZE*2 ? TILE_SIZE*2 : n - i*2;
+      }
+      else
+      {
+        valid_items_load = n - i > TILE_SIZE ? TILE_SIZE : n - i;
+        valid_items_store = n - i > TILE_SIZE ? TILE_SIZE : n - i;
+      }
+      local_abs_max = __ldg(&absmax[(i+threadIdx.x*NUM_PER_TH)/(blocksize)]);
 
       __syncthreads();
-      LoadChar(loadchar).Load(&(A[i]), qvals, valid_items, 128);
+      LoadChar(loadchar).Load(&(A[i]), qvals, valid_items_load, 128);
 
-      // load code through read-only cache via __ldg
+
-      #pragma unroll NUM_PER_TH
+      if(FP4)
-      for(int j = 0; j < NUM_PER_TH; j++)
+      {
-        vals[j] = __ldg(&code[qvals[j]])*local_abs_max;
+        #pragma unroll NUM_PER_TH
+        for(int j = 0; j < NUM_PER_TH; j++)
+        {
+          vals[j*2] = dDequantizeFP4(qvals[j] >> 4, local_abs_max*0.083333f);
+          vals[j*2 + 1] = dDequantizeFP4(qvals[j] & 0x0F, local_abs_max*0.083333);
+        }
+      }
+      else
+      {
+        // load code through read-only cache via __ldg
+        #pragma unroll NUM_PER_TH
+        for(int j = 0; j < NUM_PER_TH; j++)
+          vals[j] = __ldg(&code[qvals[j]])*local_abs_max;
+      }
 
       __syncthreads();
-      StoreT(storet).Store(&(out[i]), vals, valid_items);
+      StoreT(storet).Store(&(out[FP4 ? i*2 : i]), vals, valid_items_store);
   }
 }
 
+//template<typename T, int BLOCK_SIZE, int THREADS, int NUM_PER_TH, int TILE_SIZE>
+//__global__ void kDequantizeBlockwiseFP4(unsigned char * A, float * absmax, T *out, const int n_store)
+//{
+//
+//  const int n_load = n_store/2;
+//  const int base_idx = (blockIdx.x * TILE_SIZE);
+//
+//  T vals[NUM_PER_TH*2];
+//  unsigned char qvals[NUM_PER_TH];
+//
+//  int valid_items = (base_idx + TILE_SIZE) > n_load ? ((base_idx+TILE_SIZE) - n_load) : TILE_SIZE;
+//  int idx = base_idx + (threadIdx.x*NUM_PER_TH);
+//
+//  float local_abs_max = __ldg(&absmax[idx/BLOCK_SIZE]);
+//
+//  if(valid_items == TILE_SIZE)
+//  {
+//    // we do 64 byte loads so we can 128 byte stores
+//    reinterpret_cast<int2(&)[NUM_PER_THREAD/8]>(qvals)[0] = reinterpret_cast<int2*>(A)[idx/8];
+//  }
+//  else
+//  {
+//    #pragma unroll
+//    for(int j = 0; j < NUM_PER_TH; j++)
+//      if(idx+j < n_load)
+//        qvals[j] = A[idx+j];
+//      else
+//        qvals[j] = 0;
+//  }
+//
+//
+//  #pragma unroll NUM_PER_TH
+//  for(int j = 0; j < NUM_PER_TH; j++)
+//  {
+//    vals[j*2] = dDequantizeFP4(qvals[j] & 0xF0, local_abs_max*12.0f);
+//    vals[j*2 + 1] = dDequantizeFP4(qvals[j] & 0x0F, local_abs_max*12.0f);
+//  }
+//
+//
+//  reinterpret_cast<int4(&)[NUM_PER_THREAD/8]>(qvals)[0] = reinterpret_cast<int4*>(A)[idx/8];
+//  reinterpret_cast<int4*>(A)[idx/16] = reinterpret_cast<int4(&)[16]>(local_valC)[j/num_items];
+//
+//
+//}
+
 
 __global__ void kDequantize(float *code, unsigned char *A, float *out, const int n)
 {
@@ -2523,7 +2674,6 @@ __global__ void kspmm_coo_very_sparse_naive(int *max_count, int *max_idx, int *o
           // 4. Multiply the tile -> accumulate outputs in shared memory until 128 bytes it reached
           int idx = idx_col_B + (warp_idx*SPMM_ITEMS) + j;
           if(idx >= colsB){ break; }
-          //printf("%i %i\n", (row_offset+idx) % num_items, row_offset+idx);
           if((idx+num_items < colsB))
           {
             if(BITS == 8)
@@ -2543,8 +2693,6 @@ __global__ void kspmm_coo_very_sparse_naive(int *max_count, int *max_idx, int *o
           #pragma unroll num_items
           for(int k = 0; k < num_items; k++)
           {
-            //if((float)local_valsB[k] != 0.0)
-            //  printf("%f %i %i %i\n", (float)local_valsB[k], k, idx, colsB);
             if(BITS == 8 && dequant_stats != NULL)
               // we do texture cache reads (__ldg) on dequant_stats which should be super fast
             {
@@ -2789,38 +2937,42 @@ MAKE_optimizerStatic8bit2State(ADAM, float)
 template __global__ void kPercentileClipping<float, 2048, 4>(float * __restrict__ g, float *gnorm_vec, int step, const int n);
 template __global__ void kPercentileClipping<half, 2048, 4>(half * __restrict__ g, float *gnorm_vec, int step, const int n);
 
-template __global__ void kQuantizeBlockwise<half, 4096, 4, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 4096, 4, 0, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 4096, 4, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 4096, 4, 0, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<half, 4096, 4, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 4096, 4, 1, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 4096, 4, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 4096, 4, 1, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<half, 2048, 4, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 2048, 4, 0, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 2048, 4, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 2048, 4, 0, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<half, 1024, 4, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 1024, 4, 0, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 1024, 4, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 1024, 4, 0, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<half, 512, 2, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 512, 2, 0, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 512, 2, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 512, 2, 0, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<half, 256, 2, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 256, 2, 0, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 256, 2, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 256, 2, 0, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<half, 128, 2, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 128, 2, 0, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 128, 2, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 128, 2, 0, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<half, 64, 1, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<half, 64, 2, 0, 0>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kQuantizeBlockwise<float, 64, 1, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template __global__ void kQuantizeBlockwise<float, 64, 2, 0, 0>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
 
-template __global__ void kDequantizeBlockwise<half, 4096, 1024, 4>(float *code, unsigned char * A, float * absmax, half *out, const int n);
+template __global__ void kQuantizeBlockwise<half, 4096, 4, 0, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<float, 4096, 1024, 4>(float *code, unsigned char * A, float * absmax, float *out, const int n);
+template __global__ void kQuantizeBlockwise<float, 4096, 4, 0, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<half, 2048, 512, 4>(float *code, unsigned char * A, float * absmax, half *out, const int n);
+template __global__ void kQuantizeBlockwise<half, 2048, 4, 0, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<float, 2048, 512, 4>(float *code, unsigned char * A, float * absmax, float *out, const int n);
+template __global__ void kQuantizeBlockwise<float, 2048, 4, 0, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<half, 1024, 256, 4>(float *code, unsigned char * A, float * absmax, half *out, const int n);
+template __global__ void kQuantizeBlockwise<half, 1024, 4, 0, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<float, 1024, 256, 4>(float *code, unsigned char * A, float * absmax, float *out, const int n);
+template __global__ void kQuantizeBlockwise<float, 1024, 4, 0, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<half, 512, 256, 2>(float *code, unsigned char * A, float * absmax, half *out, const int n);
+template __global__ void kQuantizeBlockwise<half, 512, 2, 0, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<float, 512, 256, 2>(float *code, unsigned char * A, float * absmax, float *out, const int n);
+template __global__ void kQuantizeBlockwise<float, 512, 2, 0, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<half, 256, 128, 2>(float *code, unsigned char * A, float * absmax, half *out, const int n);
+template __global__ void kQuantizeBlockwise<half, 256, 2, 0, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<float, 256, 128, 2>(float *code, unsigned char * A, float * absmax, float *out, const int n);
+template __global__ void kQuantizeBlockwise<float, 256, 2, 0, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<half, 128, 64, 2>(float *code, unsigned char * A, float * absmax, half *out, const int n);
+template __global__ void kQuantizeBlockwise<half, 128, 2, 0, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<float, 128, 64, 2>(float *code, unsigned char * A, float * absmax, float *out, const int n);
+template __global__ void kQuantizeBlockwise<float, 128, 2, 0, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<half, 64, 64, 1>(float *code, unsigned char * A, float * absmax, half *out, const int n);
+template __global__ void kQuantizeBlockwise<half, 64, 2, 0, 1>(float * code, half * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template __global__ void kDequantizeBlockwise<float, 64, 64, 1>(float *code, unsigned char * A, float * absmax, float *out, const int n);
+template __global__ void kQuantizeBlockwise<float, 64, 2, 0, 1>(float * code, float * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
 
+template __global__ void kDequantizeBlockwise<half, 512, 64, 8, 1>(float *code, unsigned char * A, float * absmax, half *out, const int blocksize, const int n);
+template __global__ void kDequantizeBlockwise<float, 512, 64, 8, 1>(float *code, unsigned char * A, float * absmax, float *out, const int blocksize, const int n);
+template __global__ void kDequantizeBlockwise<half, 512, 64, 8, 0>(float *code, unsigned char * A, float * absmax, half *out, const int blocksize, const int n);
+template __global__ void kDequantizeBlockwise<float, 512, 64, 8, 0>(float *code, unsigned char * A, float * absmax, float *out, const int blocksize, const int n);
 
 
 #define MAKE_OptimizerStatic8bit2StateBlockwise(oname, gtype, block_size, num_per_thread) \

csrc/kernels.cuh

@@ -14,8 +14,8 @@ template<typename T>__global__ void kEstimateQuantiles(T *__restrict__ const A,
 __global__ void kQuantize(float * code, float * __restrict__ const A, unsigned char *out, const int n);
 __global__ void kDequantize(float *code, unsigned char *A, float *out, const int n);
 
-template<typename T, int BLOCK_SIZE, int NUM_PER_TH, int STOCHASTIC> __global__ void kQuantizeBlockwise(float * code, T * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
+template<typename T, int BLOCK_SIZE, int NUM_PER_TH, int STOCHASTIC, int FP4> __global__ void kQuantizeBlockwise(float * code, T * __restrict__ const A, float *absmax, unsigned char *out, float * __restrict__ const rand, const int rand_offset, const int n);
-template<typename T, int BLOCK_SIZE, int THREADS, int NUM_PER_TH> __global__ void kDequantizeBlockwise(float *code, unsigned char * A, float * absmax, T *out, const int n);
+template<typename T, int BLOCK_SIZE, int THREADS, int NUM_PER_TH, int FP4> __global__ void kDequantizeBlockwise(float *code, unsigned char * A, float * absmax, T *out, const int blocksize, const int n);
 
 template<typename T, int OPTIMIZER, int BLOCK_SIZE, int NUM_VALS>
 __global__ void kPreconditionOptimizer32bit2State(T* g, T* p,

csrc/ops.cu

@@ -50,7 +50,7 @@ void dequantize(float *code, unsigned char *A, float *out, int n)
   CUDA_CHECK_RETURN(cudaPeekAtLastError());
 }
 
-template <typename T, int STOCHASTIC> void quantizeBlockwise(float * code, T *A, float *absmax, unsigned char *out, float *rand, int rand_offset, int blocksize, const int n)
+template <typename T, int STOCHASTIC, int FP4> void quantizeBlockwise(float * code, T *A, float *absmax, unsigned char *out, float *rand, int rand_offset, int blocksize, const int n)
 {
   int num_blocks = n/blocksize;
   num_blocks = n % blocksize == 0 ? num_blocks : num_blocks + 1;
@@ -58,42 +58,34 @@ template <typename T, int STOCHASTIC> void quantizeBlockwise(float * code, T *A,
     assert(blocksize == 4096);
 
   if(blocksize == 4096)
-    kQuantizeBlockwise<T, 4096, 4, STOCHASTIC><<<num_blocks, 1024>>>(code, A, absmax, out, rand, rand_offset, n);
+    kQuantizeBlockwise<T, 4096, 4, STOCHASTIC, 0><<<num_blocks, 1024>>>(code, A, absmax, out, rand, rand_offset, n);
   else if(blocksize == 2048)
-    kQuantizeBlockwise<T, 2048, 4, 0><<<num_blocks, 512>>>(code, A, absmax, out, rand, rand_offset, n);
+    kQuantizeBlockwise<T, 2048, 4, 0, FP4><<<num_blocks, 512>>>(code, A, absmax, out, rand, rand_offset, n);
   else if(blocksize == 1024)
-    kQuantizeBlockwise<T, 1024, 4, 0><<<num_blocks, 256>>>(code, A, absmax, out, rand, rand_offset, n);
+    kQuantizeBlockwise<T, 1024, 4, 0, FP4><<<num_blocks, 256>>>(code, A, absmax, out, rand, rand_offset, n);
   else if(blocksize == 512)
-    kQuantizeBlockwise<T, 512, 2, 0><<<num_blocks, 256>>>(code, A, absmax, out, rand, rand_offset, n);
+    kQuantizeBlockwise<T, 512, 2, 0, FP4><<<num_blocks, 256>>>(code, A, absmax, out, rand, rand_offset, n);
   else if(blocksize == 256)
-    kQuantizeBlockwise<T, 256, 2, 0><<<num_blocks, 128>>>(code, A, absmax, out, rand, rand_offset, n);
+    kQuantizeBlockwise<T, 256, 2, 0, FP4><<<num_blocks, 128>>>(code, A, absmax, out, rand, rand_offset, n);
   else if(blocksize == 128)
-    kQuantizeBlockwise<T, 128, 2, 0><<<num_blocks, 64>>>(code, A, absmax, out, rand, rand_offset, n);
+    kQuantizeBlockwise<T, 128, 2, 0, FP4><<<num_blocks, 64>>>(code, A, absmax, out, rand, rand_offset, n);
   else if(blocksize == 64)
-    kQuantizeBlockwise<T, 64, 1, 0><<<num_blocks, 64>>>(code, A, absmax, out, rand, rand_offset, n);
+    kQuantizeBlockwise<T, 64, 2, 0, FP4><<<num_blocks, 32>>>(code, A, absmax, out, rand, rand_offset, n);
 
 
   CUDA_CHECK_RETURN(cudaPeekAtLastError());
 }
 
-template<typename T> void dequantizeBlockwise(float *code, unsigned char *A, float *absmax, T *out, int blocksize, const int n)
+template<typename T, int FP4> void dequantizeBlockwise(float *code, unsigned char *A, float *absmax, T *out, int blocksize, const int n)
 {
   int num_blocks = n/blocksize;
   num_blocks = n % blocksize == 0 ? num_blocks : num_blocks + 1;
-  if(blocksize == 4096)
+  int tile_size = FP4 ? 1024 : 512;
-    kDequantizeBlockwise<T, 4096, 1024, 4><<<num_blocks, 4096/4>>>(code, A, absmax, out, n);
+
-  else if(blocksize == 2048)
+  if(FP4)
-    kDequantizeBlockwise<T, 2048, 512, 4><<<num_blocks, 2048/4>>>(code, A, absmax, out, n);
+    kDequantizeBlockwise<T, 512, 64, 8, FP4><<<(n+tile_size-1)/tile_size, 64>>>(code, A, absmax, out, blocksize/2, n);
-  else if(blocksize == 1024)
+  else
-    kDequantizeBlockwise<T, 1024, 256, 4><<<num_blocks, 1024/4>>>(code, A, absmax, out, n);
+    kDequantizeBlockwise<T, 512, 64, 8, FP4><<<(n+tile_size-1)/tile_size, 64>>>(code, A, absmax, out, blocksize, n);
-  else if(blocksize == 512)
-    kDequantizeBlockwise<T, 512, 256, 2><<<num_blocks, 512/2>>>(code, A, absmax, out, n);
-  else if(blocksize == 256)
-    kDequantizeBlockwise<T, 256, 128, 2><<<num_blocks, 256/2>>>(code, A, absmax, out, n);
-  else if(blocksize == 128)
-    kDequantizeBlockwise<T, 128, 64, 2><<<num_blocks, 128/2>>>(code, A, absmax, out, n);
-  else if(blocksize == 64)
-    kDequantizeBlockwise<T, 64, 64, 1><<<num_blocks, 64/1>>>(code, A, absmax, out, n);
 
   CUDA_CHECK_RETURN(cudaPeekAtLastError());
 }
@@ -688,12 +680,16 @@ template void transformRowToFormat<COL_AMPERE, 1>(char * A, char *out, int rows,
 template void estimateQuantiles(half *A, float *code, float offset, int n);
 template void estimateQuantiles(float *A, float *code, float offset, int n);
 
-template void quantizeBlockwise<half, 0>(float * code, half *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
+template void quantizeBlockwise<half, 0, 0>(float * code, half *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
-template void quantizeBlockwise<float, 0>(float * code, float *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
+template void quantizeBlockwise<float, 0, 0>(float * code, float *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
-template void quantizeBlockwise<half, 1>(float * code, half *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
+template void quantizeBlockwise<half, 0, 1>(float * code, half *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
-template void quantizeBlockwise<float, 1>(float * code, float *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
+template void quantizeBlockwise<float, 0, 1>(float * code, float *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
-template void dequantizeBlockwise<half>(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n);
+template void quantizeBlockwise<half, 1, 0>(float * code, half *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
-template void dequantizeBlockwise<float>(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n);
+template void quantizeBlockwise<float, 1, 0>(float * code, float *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
+template void dequantizeBlockwise<half, 0>(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n);
+template void dequantizeBlockwise<float, 0>(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n);
+template void dequantizeBlockwise<half, 1>(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n);
+template void dequantizeBlockwise<float, 1>(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n);
 
 #define MAKE_optimizer32bit(name, gtype) \
 template void optimizer32bit<gtype, name>(gtype* g, gtype* p, \

csrc/ops.cuh

@@ -128,8 +128,8 @@ template <typename T> void estimateQuantiles(T *A, float *code, float offset, in
 
 void quantize(float *code, float *A, unsigned char *out, int n);
 void dequantize(float *code, unsigned char *A, float *out, int n);
-template <typename T, int STOCHASTIC> void quantizeBlockwise(float * code, T *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
+template <typename T, int STOCHASTIC, int FP4> void quantizeBlockwise(float * code, T *A, float *absmax, unsigned char *out, float* rand, int rand_offset, int blocksize, const int n);
-template<typename T> void dequantizeBlockwise(float *code, unsigned char *A, float *absmax, T *out, int block_size, const int n);
+template<typename T, int FP4> void dequantizeBlockwise(float *code, unsigned char *A, float *absmax, T *out, int block_size, const int n);
 
 template<typename T, int OPTIMIZER> void optimizer32bit(T* g, T* p,
                 float* state1, float* state2, float *unorm, float max_unorm, float param_norm,

csrc/pythonInterface.c

@@ -75,13 +75,17 @@ MAKE_BLOCKWISE8(adagrad, ADAGRAD, float, 32)
 void percentileClipping_g32(float * g, float *gnorm_vec, int step, const int n){ percentileClipping<float>(g, gnorm_vec, step, n); }
 void percentileClipping_g16(half * g, float *gnorm_vec, int step, const int n){ percentileClipping<half>(g, gnorm_vec, step, n); }
 
-void quantizeBlockwise_fp16(float * code, half *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise<half, 0>(code, A, absmax, out, NULL, 0, blocksize, n); }
+void quantizeBlockwise_fp16(float * code, half *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise<half, 0, 0>(code, A, absmax, out, NULL, 0, blocksize, n); }
-void quantizeBlockwise_fp32(float * code, float *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise<float, 0>(code, A, absmax, out, NULL, 0, blocksize, n); }
+void quantizeBlockwise_fp32(float * code, float *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise<float, 0, 0>(code, A, absmax, out, NULL, 0, blocksize, n); }
-void quantizeBlockwise_stochastic_fp16(float * code, half *A, float *absmax, unsigned char *out, float* rand, int rand_offset, const int n){ quantizeBlockwise<half, 1>(code, A, absmax, out, rand, rand_offset, 4096, n); }
+void quantizeBlockwise_stochastic_fp16(float * code, half *A, float *absmax, unsigned char *out, float* rand, int rand_offset, const int n){ quantizeBlockwise<half, 1, 0>(code, A, absmax, out, rand, rand_offset, 4096, n); }
-void quantizeBlockwise_stochastic_fp32(float * code, float *A, float *absmax, unsigned char *out, float* rand, int rand_offset, const int n){ quantizeBlockwise<float, 1>(code, A, absmax, out, rand, rand_offset, 4096, n); }
+void quantizeBlockwise_stochastic_fp32(float * code, float *A, float *absmax, unsigned char *out, float* rand, int rand_offset, const int n){ quantizeBlockwise<float, 1, 0>(code, A, absmax, out, rand, rand_offset, 4096, n); }
+void quantizeBlockwise_fp16_fp4(float * code, half *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise<half, 0, 1>(NULL, A, absmax, out, NULL, 0, blocksize, n); }
+void quantizeBlockwise_fp32_fp4(float * code, float *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise<float, 0, 1>(NULL, A, absmax, out, NULL, 0, blocksize, n); }
 
-void dequantizeBlockwise_fp16(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n){ dequantizeBlockwise<half>(code, A, absmax, out, blocksize, n); } \
+void dequantizeBlockwise_fp16(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n){ dequantizeBlockwise<half, 0>(code, A, absmax, out, blocksize, n); } \
-void dequantizeBlockwise_fp32(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n){ dequantizeBlockwise<float>(code, A, absmax, out, blocksize, n); }
+void dequantizeBlockwise_fp32(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n){ dequantizeBlockwise<float, 0>(code, A, absmax, out, blocksize, n); }
+void dequantizeBlockwise_fp16_fp4(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n){ dequantizeBlockwise<half, 1>(NULL, A, absmax, out, blocksize, n); } \
+void dequantizeBlockwise_fp32_fp4(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n){ dequantizeBlockwise<float, 1>(NULL, A, absmax, out, blocksize, n); }
 
 #define MAKE_FUNC_TRANSFORM(fbits, fsrc, ftrgt, ftranspose, dtype, src, target, transpose, bits) \
 void transform_##fbits##_##fsrc##_to_##ftrgt##_##ftranspose(cublasLtHandle_t ltHandle, dtype *A, dtype *out, int dim1, int dim2) \
@@ -148,6 +152,11 @@ extern "C"
   void cdequantize_blockwise_fp16(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n){ dequantizeBlockwise_fp16(code, A, absmax, out, blocksize, n); }
   void cdequantize_blockwise_fp32(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n){ dequantizeBlockwise_fp32(code, A, absmax, out, blocksize, n); }
 
+  void cquantize_blockwise_fp16_fp4(float * code, half *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise_fp16_fp4(code, A, absmax, out, blocksize, n); }
+  void cquantize_blockwise_fp32_fp4(float * code, float *A, float *absmax, unsigned char *out, int blocksize, const int n){ quantizeBlockwise_fp32_fp4(code, A, absmax, out, blocksize, n); }
+  void cdequantize_blockwise_fp16_fp4(float *code, unsigned char *A, float *absmax, half *out, int blocksize, const int n){ dequantizeBlockwise_fp16_fp4(code, A, absmax, out, blocksize, n); }
+  void cdequantize_blockwise_fp32_fp4(float *code, unsigned char *A, float *absmax, float *out, int blocksize, const int n){ dequantizeBlockwise_fp32_fp4(code, A, absmax, out, blocksize, n); }
+
 	#define MAKE_CFUNC32(name, gtype, gbits) \
 	void c##name##32bit_g##gbits(gtype *g, gtype *p, \
 								 float* state1, float* state2, float *unorm, float max_unorm, float param_norm, \

tests/test_functional.py

@@ -152,7 +152,7 @@ def test_dynamic_quantization():
 
 def test_dynamic_blockwise_quantization():
     #print('')
-    for blocksize in [4096, 2048, 1024, 512]:
+    for blocksize in [4096, 2048, 1024, 512, 256, 128, 64]:
         diffs = []
         reldiffs = []
         for i in range(100):
@@ -2189,7 +2189,88 @@ def test_bench_dequantization():
     torch.cuda.synchronize()
     t0 = time.time()
     for i in range(100):
-        F.dequantize_blockwise(qa, SA, blocksize=2048)
+        #F.dequantize_blockwise(qa, SA, blocksize=2048)
+        qa, SA = F.quantize_blockwise(a)
     torch.cuda.synchronize()
     #print((time.time()-t0)/1e6)
 
+
+
+def test_fp4_quant():
+    vals = list(product([0, 1], repeat=4))
+
+    code = {}
+    for bits in vals:
+        result = 0
+        bias = 3
+        sign, e1, e2, p1 = bits
+        idx = sign*8 + e1*4 + e2*2 + p1*1
+        sign = -1.0 if sign else 1.0
+        exp = e1*2 + e2*1
+        if exp == 0:
+            # sub-normal
+            if p1 == 0: result = 0
+            else: result = sign*0.0625
+        else:
+            # normal
+            exp = 2**(-exp + bias + 1)
+            frac = 1.5 if p1 else 1.0
+            result = sign*exp*frac
+        code[idx] = result
+
+    A1 = torch.randn(1024, 1024, device='cuda').half()
+    qa, SA = F.quantize_fp4(A1, blocksize=64)
+    A2 = F.dequantize_fp4(qa, SA)
+    #qa, SA = F.quantize_fp4(A1, blocksize=128)
+    #A2 = F.dequantize_fp4(qa, SA, blocksize=128)
+
+    #A1 = A1.flatten().sort()[0]
+    #A2 = A2.flatten().sort()[0]
+
+    #print(A1)
+    #print(A2)
+
+    err = (A1 - A2).abs().float()
+    relerr = (err/A1.abs().float()).mean()
+    err = err.mean()
+
+    print(err, relerr)
+
+
+
+
+    #assert err.item() < 0.1
+    #assert relerr.item() < 0.28
+
+
+def test_bench_fp4_dequant():
+    blocksize = 256
+    a = torch.rand(1024*12*4, 1024*12, device='cuda').half()
+    qa, SA = F.quantize_fp4(a, blocksize=blocksize)
+
+    input_size = a.numel()/2
+    output_size = a.numel()*2
+    num_bytes = input_size+output_size
+    GB = num_bytes/1e9
+    max_theoretical_s =  GB/768
+    print(max_theoretical_s*1e6)
+    b = torch.randn(128, 1024*12, device='cuda').half()
+
+    iters = 5
+    torch.cuda.synchronize()
+    t0 = time.time()
+    for i in range(iters):
+        F.dequantize_fp4(qa, SA, blocksize=blocksize)
+        #b.copy_(a)
+    torch.cuda.synchronize()
+    print((time.time()-t0)/iters*1e6)
+
+    torch.cuda.synchronize()
+    t0 = time.time()
+    for i in range(iters):
+        torch.matmul(b, a.t())
+    torch.cuda.synchronize()
+    print((time.time()-t0)/iters*1e6)
+
+
+